Merge pull request 'fix: resolve skill audit findings' (#97) from fix/skill-audit-compliance into main

2026-09-11 19:47:12 +03:00 · 2026-07-11 09:25:51 -04:00
parent 61bacbfaf1 a5f7e74972
commit 0ab926b57b
76 changed files with 716 additions and 447 deletions
@@ -78,3 +78,23 @@ The synthesis report includes:
 ## License
 MIT
 ## Why Install This Skill
 This skill packages practical, reusable guidance for this domain so you can move from a real task to a dependable result without rebuilding the workflow each time.
 ## What You Get
 A focused workflow in SKILL.md, with the referenced scripts, templates, and supporting material available when the task needs them.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
 ## Requirements
 Check the compatibility requirements in SKILL.md before using commands or integrations from this skill.
@@ -1,22 +1,18 @@
 ---
 name: arr-cli
-description: >-
+description: Manage your Radarr (movies) and Sonarr (TV series) media library from
-  Manage your Radarr (movies) and Sonarr (TV series) media library from the terminal.
+  the terminal. Search and browse movies/series, add new content, check calendars,
-  Search and browse movies/series, add new content, check calendars, view queue and
+  view queue and download history, inspect quality profiles, and trigger searches.
-  download history, inspect quality profiles, and trigger searches. Use when the user
+  Use when the user mentions Radarr, Sonarr, the *arr stack, adding a movie or series,
-  mentions Radarr, Sonarr, the *arr stack, adding a movie or series, checking the
+  checking the library, finding something to watch, what's in the queue, download
-  library, finding something to watch, what's in the queue, download history,
+  history, upcoming releases, media server setup, or movie/TV automation.
  upcoming releases, media server setup, or movie/TV automation.
 license: MIT
-compatibility: >-
+compatibility: Python 3.8+ with `requests` library. Requires ARR_SERVER_RADARR and
-  Python 3.8+ with `requests` library. Requires ARR_SERVER_RADARR and ARR_KEY_RADARR
+  ARR_KEY_RADARR (for Radarr) or ARR_SERVER_SONARR and ARR_KEY_SONARR (for Sonarr).
-  (for Radarr) or ARR_SERVER_SONARR and ARR_KEY_SONARR (for Sonarr). API keys from
+  API keys from each app's Settings → General.
  each app's Settings → General.
 metadata:
-  tags: [radarr, sonarr, arr-stack, media-server, movie-automation, tv-series, api-client]
+  tags: radarr, sonarr, arr-stack, media-server, movie-automation, tv-series, api-client
-  sources:
+  sources: https://radarr.video/, https://sonarr.tv/
    - https://radarr.video/
    - https://sonarr.tv/
 ---
 # arr-cli — Radarr + Sonarr Media Library Management
@@ -30,3 +30,8 @@ AutoGen differs from other frameworks in the portfolio: it's conversation-driven
 ## Requirements
 Python 3.8+ with `autogen-agentchat` and `autogen-ext` packages.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -28,3 +28,8 @@ Load this when you need a brand card for press/stakeholders, visual identity doc
 ## Requirements
 Python 3.8+ for the CLI; image generation requires a compatible backend. Templates are plain markdown.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -38,3 +38,8 @@ Load this when you hear "Tailscale," "Headscale," "tailnet," "mesh VPN," "WireGu
 ## Requirements
 Bash, Python 3.8+, jq, curl. Access to a Headscale server or the `headscale` CLI. Tailscale client on target machines.
 ## Why Install This Skill
 This skill packages practical, reusable guidance for this domain so you can move from a real task to a dependable result without rebuilding the workflow each time.
@@ -1,16 +1,16 @@
 ---
 name: tailscale
-description: >-
+description: 'Self-hosted Tailscale/Headscale ecosystem: deploy and manage a Headscale
-  Self-hosted Tailscale/Headscale ecosystem: deploy and manage a Headscale control server,
+  control server, configure tailscale clients, manage ACL policies, node lifecycle,
-  configure tailscale clients, manage ACL policies, node lifecycle, subnet routing, DERP relays,
+  subnet routing, DERP relays, and backup/migration. Use when the user mentions Tailscale,
-  and backup/migration. Use when the user mentions Tailscale, Headscale, tailnet, mesh VPN,
+  Headscale, tailnet, mesh VPN, WireGuard mesh, or self-hosted VPN infrastructure.'
  WireGuard mesh, or self-hosted VPN infrastructure.
 license: MIT
-compatibility: Requires bash, Python 3.8+, jq, curl, and access to a Headscale server or
+compatibility: Requires bash, Python 3.8+, jq, curl, and access to a Headscale server
-  the `headscale` CLI. Tailscale client (`tailscale`) must be installed on target machines.
+  or the `headscale` CLI. Tailscale client (`tailscale`) must be installed on target
  machines.
 metadata:
-  tags: [tailscale, headscale, vpn, wireguard, mesh, networking, homelab]
+  tags: tailscale, headscale, vpn, wireguard, mesh, networking, homelab
-  spec-version: "1.0"
+  spec-version: '1.0'
 ---
 # Tailscale + Headscale Skill Bundle
@@ -1,7 +1,10 @@
 ---
 name: headscale-backup
-description: Backup, restore, and migrate Headscale installations — SQLite database, configuration file, policy file, and TLS certificates. Use when backing up a headscale server before upgrades, migrating to new hardware, or restoring from a disaster.
+description: Backup, restore, and migrate Headscale installations — SQLite database,
-category: devops
+  configuration file, policy file, and TLS certificates. Use when backing up a headscale
  server before upgrades, migrating to new hardware, or restoring from a disaster.
 metadata:
  category: devops
 ---
 # headscale-backup
@@ -1,17 +1,13 @@
 ---
 name: headscale-deploy
-description: Deploy, configure, and maintain a self-hosted Headscale control server on Linux or Docker. Use when setting up a new Headscale instance, troubleshooting deployment issues, or configuring server settings.
+description: Deploy, configure, and maintain a self-hosted Headscale control server
  on Linux or Docker. Use when setting up a new Headscale instance, troubleshooting
  deployment issues, or configuring server settings.
 license: MIT
 compatibility: linux, docker
 metadata:
-  tags:
+  tags: headscale, tailscale, wireguard, vpn, deployment, devops
-    - headscale
+  spec-version: '1.0'
    - tailscale
    - wireguard
    - vpn
    - deployment
    - devops
  spec-version: 1.0
 ---
 # headscale-deploy
@@ -1,7 +1,11 @@
 ---
 name: headscale-derp
-description: Configure and manage DERP relay servers for Tailscale/Headscale — embedded and standalone DERP deployment, latency testing, and connectivity diagnostics. Use when direct peer connections fail, traffic is routed through DERP relays, or setting up custom relay regions.
+description: Configure and manage DERP relay servers for Tailscale/Headscale — embedded
-category: devops
+  and standalone DERP deployment, latency testing, and connectivity diagnostics. Use
  when direct peer connections fail, traffic is routed through DERP relays, or setting
  up custom relay regions.
 metadata:
  category: devops
 ---
 # headscale-derp Skill
@@ -1,7 +1,10 @@
 ---
 name: headscale-node-lifecycle
-description: Manage the full lifecycle of nodes in a Headscale tailnet — generate pre-authenticated keys, register, approve, tag, list, and decommission nodes. Use when adding new devices, generating auth keys for automation, or managing node inventory.
+description: Manage the full lifecycle of nodes in a Headscale tailnet — generate
-category: devops
+  pre-authenticated keys, register, approve, tag, list, and decommission nodes. Use
  when adding new devices, generating auth keys for automation, or managing node inventory.
 metadata:
  category: devops
 ---
 # headscale-node-lifecycle
@@ -1,7 +1,10 @@
 ---
 name: headscale-routing
-description: Configure subnet routers and exit nodes in a Headscale tailnet to extend mesh access to non-Tailscale devices and route internet traffic. Use when setting up subnet routing for LAN devices or configuring exit nodes for privacy.
+description: Configure subnet routers and exit nodes in a Headscale tailnet to extend
-category: devops
+  mesh access to non-Tailscale devices and route internet traffic. Use when setting
  up subnet routing for LAN devices or configuring exit nodes for privacy.
 metadata:
  category: devops
 ---
 # headscale-routing
@@ -1,7 +1,11 @@
 ---
 name: tailnet-policy
-description: Author, test, and deploy Tailscale-compatible huJSON policy files for Headscale tailnets — ACLs, Grants, Tags, Auto Approvers, Tailscale SSH rules. Use when configuring access control, writing policy files, or troubleshooting connectivity issues caused by ACLs.
+description: Author, test, and deploy Tailscale-compatible huJSON policy files for
-category: devops
+  Headscale tailnets — ACLs, Grants, Tags, Auto Approvers, Tailscale SSH rules. Use
  when configuring access control, writing policy files, or troubleshooting connectivity
  issues caused by ACLs.
 metadata:
  category: devops
 ---
 # tailnet-policy
@@ -1,7 +1,10 @@
 ---
 name: tailscale-client
-description: Install, configure, and troubleshoot the official Tailscale client when connected to a Headscale self-hosted control server. Use when connecting a new device, diagnosing connectivity, checking peer status, or troubleshooting DERP relay issues.
+description: Install, configure, and troubleshoot the official Tailscale client when
-category: devops
+  connected to a Headscale self-hosted control server. Use when connecting a new device,
  diagnosing connectivity, checking peer status, or troubleshooting DERP relay issues.
 metadata:
  category: devops
 ---
 # tailscale-client
@@ -36,3 +36,8 @@ Load this when you want to understand your own process, formalize a workflow, sh
 ## Requirements
 Hermes Agent (uses skill_view, memory, session context scanning, write_file for bundle generation). Output bundles are standard Agent Skills format.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,18 +1,16 @@
 ---
 name: workflow-architect
-description: >-
+description: Discover your actual workflow through conversation or observation, then
-  Discover your actual workflow through conversation or observation, then generate
+  generate a tailored skills bundle that encodes it as loadable agent skills with
-  a tailored skills bundle that encodes it as loadable agent skills with trigger
+  trigger conditions. Use when you want to understand your own process, formalize
-  conditions. Use when you want to understand your own process, formalize it, or
+  it, or share it with collaborators. Also use when a session feels aimless — this
-  share it with collaborators. Also use when a session feels aimless — this skill
+  skill gives it structure.
  gives it structure.
 license: MIT
-compatibility: >-
+compatibility: Hermes Agent — uses skill_view(), memory tool, session context scanning,
  Hermes Agent — uses skill_view(), memory tool, session context scanning,
  and write_file for bundle generation. Output bundles are standard Agent Skills.
 metadata:
-  tags: [workflow, meta, productivity, skills-bundle, onboarding, process]
+  tags: workflow, meta, productivity, skills-bundle, onboarding, process
-  spec-version: "1.0"
+  spec-version: '1.0'
 ---
 # Workflow Architect
@@ -1,15 +1,14 @@
 ---
 name: bundle-builder
-description: >-
+description: The synthesis engine for workflow-architect. Reads accumulated workflow
-  The synthesis engine for workflow-architect. Reads accumulated workflow
+  state from memory and renders it into a valid Agent Skills bundle directory with
-  state from memory and renders it into a valid Agent Skills bundle directory
+  sub-skills, manifest, decision map, and optional kanban board. Loaded after the
-  with sub-skills, manifest, decision map, and optional kanban board. Loaded
+  interviewer or observer achieves convergence.
  after the interviewer or observer achieves convergence.
 license: MIT
 compatibility: Hermes Agent — uses write_file for bundle output, shell_quote
 metadata:
-  tags: [workflow, generation, bundle, synthesis]
+  tags: workflow, generation, bundle, synthesis
-  spec-version: "1.0"
+  spec-version: '1.0'
 ---
 # Bundle Builder — Workflow Synthesis Engine
@@ -1,15 +1,14 @@
 ---
 name: interviewer
-description: >-
+description: The active interrogation sub-skill for workflow-architect. Guides the
-  The active interrogation sub-skill for workflow-architect. Guides the user
+  user through 8-15 adaptive questions to discover their workflow phases, branching
-  through 8-15 adaptive questions to discover their workflow phases, branching
+  signals, tool preferences, and friction points. Loaded by the umbrella workflow-architect
-  signals, tool preferences, and friction points. Loaded by the umbrella
+  skill when running in active mode.
  workflow-architect skill when running in active mode.
 license: MIT
 compatibility: Hermes Agent — uses memory tool for state persistence
 metadata:
-  tags: [workflow, interview, discovery, process]
+  tags: workflow, interview, discovery, process
-  spec-version: "1.0"
+  spec-version: '1.0'
 ---
 # Interviewer — Active Workflow Discovery
@@ -1,15 +1,13 @@
 ---
 name: observer
-description: >-
+description: The passive observation sub-skill for workflow-architect. Loaded silently,
-  The passive observation sub-skill for workflow-architect. Loaded silently,
+  it stays dormant until a trigger phrase activates it. On activation, it scans the
-  it stays dormant until a trigger phrase activates it. On activation, it
+  current session context to infer the user's workflow from what actually happened.
  scans the current session context to infer the user's workflow from what
  actually happened.
 license: MIT
 compatibility: Hermes Agent — uses session_search and context scanning
 metadata:
-  tags: [workflow, observation, discovery, passive]
+  tags: workflow, observation, discovery, passive
-  spec-version: "1.0"
+  spec-version: '1.0'
 ---
 # Observer — Passive Workflow Discovery
@@ -27,3 +27,8 @@ Load this when building a new CLI tool, refactoring an existing tool that causes
 ## Requirements
 Bash, Python 3.8+, jq, and a standard Unix CLI environment.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,19 +1,15 @@
 ---
 name: cli-builder
-description: >-
+description: 'Build or refactor CLI tools designed for AI agent consumption: non-interactive,
-  Build or refactor CLI tools designed for AI agent consumption: non-interactive,
+  flag-driven, idempotent, with --json output and --dry-run preview. Use when creating
-  flag-driven, idempotent, with --json output and --dry-run preview. Use when
+  a new script the agent will call, adding agent-friendly flags to an existing tool,
-  creating a new script the agent will call, adding agent-friendly flags to an
+  or debugging why an agent keeps failing to use your CLI.'
  existing tool, or debugging why an agent keeps failing to use your CLI.
 license: MIT
 compatibility: Requires bash, Python 3.8+, jq, and standard Unix CLI environment.
 metadata:
-  tags: [cli, agent-tooling, design-patterns, automation]
+  tags: cli, agent-tooling, design-patterns, automation
-  sources:
+  sources: https://x.com/ericzakariasson/status/2036762680401223946, https://www.scalekit.com/blog/mcp-vs-cli-use,
-    - https://x.com/ericzakariasson/status/2036762680401223946
+    https://github.com/ComposioHQ/awesome-agent-clis, https://ronnierocha.dev/blog/dont-build-mcps-build-cli-tools
    - https://www.scalekit.com/blog/mcp-vs-cli-use
    - https://github.com/ComposioHQ/awesome-agent-clis
    - https://ronnierocha.dev/blog/dont-build-mcps-build-cli-tools
 ---
 # CLI Builder — Agent-Friendly Tool Design
 ## Overview
@@ -484,12 +480,9 @@ The default for this repo is **`scripts/` inside the skill** — it follows the
 ### Skill Wrapper Template
 See [references/skill-wrapper-example.md](references/skill-wrapper-example.md) for a complete worked example wrapping a hypothetical API CLI, including frontmatter, essential commands, gotchas, and auth wiring documentation.
 ## Agent-Readiness Checklist
 Use [the agent-readiness checklist](references/agent-readiness-checklist.md) before shipping a CLI.
 ## References
 - [templates/bash-cli-scaffold.sh](templates/bash-cli-scaffold.sh) — Full bash project template with pre-wired global flags, logging helpers, and subcommand dispatch. Use as a starting point for any bash CLI.
@@ -28,3 +28,8 @@ Load this when inspecting ICC profiles, converting between color spaces, checkin
 ## Requirements
 ImageMagick, Exiftool, ArgyllCMS, LittleCMS (all platform-independent). Python scripts require Python 3.8+.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,21 +1,16 @@
 ---
 name: confluence-cli
-description: >-
+description: 'Interact with Atlassian Confluence from the terminal: list spaces, browse
-  Interact with Atlassian Confluence from the terminal: list spaces,
+  pages, view page content, search with CQL, and create pages. Use when the user mentions
-  browse pages, view page content, search with CQL, and create pages.
+  Confluence, a space key (e.g. DEV), or asks about documentation, wiki pages, space
-  Use when the user mentions Confluence, a space key (e.g. DEV), or
+  content, or knowledge base articles.'
  asks about documentation, wiki pages, space content, or knowledge
  base articles.
 license: MIT
-compatibility: Requires CONFLUENCE_EMAIL and CONFLUENCE_API_TOKEN env vars
+compatibility: Requires CONFLUENCE_EMAIL and CONFLUENCE_API_TOKEN env vars (free from
-  (free from id.atlassian.com/manage/api-tokens), Python 3.8+, and the
+  id.atlassian.com/manage/api-tokens), Python 3.8+, and the `requests` library. Also
-  `requests` library. Also requires CONFLUENCE_SERVER (defaults to
+  requires CONFLUENCE_SERVER (defaults to your-domain.atlassian.net).
  your-domain.atlassian.net).
 metadata:
-  tags: [confluence, atlassian, wiki, documentation, knowledge-base, api-client]
+  tags: confluence, atlassian, wiki, documentation, knowledge-base, api-client
-  sources:
+  sources: https://developer.atlassian.com/cloud/confluence/rest/v2/, https://id.atlassian.com/manage/api-tokens
    - https://developer.atlassian.com/cloud/confluence/rest/v2/
    - https://id.atlassian.com/manage/api-tokens
 ---
 # confluence-cli — Confluence Wiki from the Terminal
@@ -27,3 +27,13 @@ CrewAI is higher-abstraction than LangGraph (which is a low-level state machine)
 ## Requirements
 Python 3.8+ with `crewai` and `crewai-tools` packages.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
@@ -26,3 +26,8 @@ Load this when setting up or troubleshooting CrowdSec, or when you need an IPS/I
 ## Requirements
 Linux server for installation. Docker for containerized deployment. Standard shell tools.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,15 +1,15 @@
 ---
 name: crowdsec
-version: 0.0.2
+description: Deploy, configure, and manage CrowdSec — the open-source, collaborative
-description: >-
+  IPS/IDPS/WAF. Covers Security Engine setup (Linux, Docker), cscli hub management,
-  Deploy, configure, and manage CrowdSec — the open-source, collaborative IPS/IDPS/WAF.
+  remediation components, AppSec WAF, profiles, notifications, blocklists, CTI, and
-  Covers Security Engine setup (Linux, Docker), cscli hub management, remediation
+  metrics. Use when setting up or troubleshooting CrowdSec.
  components, AppSec WAF, profiles, notifications, blocklists, CTI, and metrics.
  Use when setting up or troubleshooting CrowdSec.
 license: MIT
-compatibility: Any agent supporting Agent Skills format — commands use standard shell and CLI tools
+compatibility: Any agent supporting Agent Skills format — commands use standard shell
  and CLI tools
 metadata:
  source: https://docs.crowdsec.net
  version: 0.0.2
 ---
 # CrowdSec Skill
 CrowdSec is an open-source, collaborative security engine that detects and blocks malicious actors. It analyzes logs and HTTP requests using behavior-based patterns (scenarios) and enforces blocks through remediation components (bouncers).
@@ -479,12 +479,9 @@ api:
        - bouncers_ou
 ```
 Client auth types: `NoClientCert`, `VerifyClientCertIfGiven` (default), `RequireAndVerifyClientCert`
 ## References
 Load the following reference files for deeper coverage of specific topics:
 | Reference | Load when | File |
 |-----------|-----------|------|
 | Full config.yaml reference | You need every configuration directive explained | `references/config-reference.md` |
@@ -27,3 +27,8 @@ Load this when your data pipelines are growing out of control, teams disagree on
 ## Requirements
 No special system requirements. Designed for agentic AI assistants.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -86,3 +86,18 @@ All scripts default to Python computation. The `--engine r` flag outputs equival
 ## License
 MIT
 ## Why Install This Skill
 This skill packages practical, reusable guidance for this domain so you can move from a real task to a dependable result without rebuilding the workflow each time.
 ## What You Get
 A focused workflow in SKILL.md, with the referenced scripts, templates, and supporting material available when the task needs them.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,22 +1,21 @@
 ---
 name: data-scientist
-description: >-
+description: PhD-level expertise in data science, statistics, and machine learning.
-  PhD-level expertise in data science, statistics, and machine learning. Use when
+  Use when the task requires rigorous statistical analysis, experimental design, causal
-  the task requires rigorous statistical analysis, experimental design, causal
+  inference, advanced modeling, research methodology, or data science project leadership.
-  inference, advanced modeling, research methodology, or data science project
+  Load when the user asks about statistical methods, experimental design, model selection,
-  leadership. Load when the user asks about statistical methods, experimental design,
+  A/B testing, hypothesis testing, power analysis, regression, causality, Bayesian
-  model selection, A/B testing, hypothesis testing, power analysis, regression,
+  analysis, or research methodology.
  causality, Bayesian analysis, or research methodology.
 license: MIT
-compatibility: >-
+compatibility: Python 3.10+ with scipy, statsmodels, scikit-learn, pandas, numpy.
-  Python 3.10+ with scipy, statsmodels, scikit-learn, pandas, numpy. PyTorch
+  PyTorch and sklearn are the primary ML frameworks. Hardware-aware via detect-compute.py.
-  and sklearn are the primary ML frameworks. Hardware-aware via detect-compute.py.
+  Optional R engine via rpy2. Deep learning assumes NVIDIA GPU with CUDA or Apple
-  Optional R engine via rpy2. Deep learning assumes NVIDIA GPU with CUDA or Apple MPS.
+  MPS.
 metadata:
-  spec-version: "1.0"
+  spec-version: '1.0'
-  skills: [research-methodology, statistics, machine-learning, causal-inference,
+  skills: research-methodology, statistics, machine-learning, causal-inference, bayesian-analysis,
-           bayesian-analysis, experimental-design]
+    experimental-design
-  requires-toolsets: [terminal]
+  requires-toolsets: terminal
 ---
 # PhD-Level Data Science
@@ -26,3 +26,13 @@ DSPy is **not** a chain or RAG framework. It operates at the compiler layer —
 ## Requirements
 Python 3.8+ with `dspy` package.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
@@ -30,3 +30,8 @@ Load this when you encounter EPUB files — to read, create, edit, extract, conv
 ## Requirements
 Python 3.8+ with `EbookLib`. Optional: `beautifulsoup4`, `epublib`, Java (for EPUBCheck), and an LLM endpoint for knowledge extraction.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,22 +1,20 @@
 ---
 name: epub
-description: >-
+description: EPUB file format expert — read, write, and edit EPUB2/EPUB3 ebooks. Extract
-  EPUB file format expert — read, write, and edit EPUB2/EPUB3 ebooks. Extract text,
+  text, metadata, structure, and knowledge from EPUB files for enrichment or memory.
-  metadata, structure, and knowledge from EPUB files for enrichment or memory. Create
+  Create valid EPUBs from scratch. Validate against the EPUB specification. Use when
-  valid EPUBs from scratch. Validate against the EPUB specification. Use when the user
+  the user mentions epub, ebook, EPUB file, ebook format, read epub, write epub, create
-  mentions epub, ebook, EPUB file, ebook format, read epub, write epub, create ebook,
+  ebook, extract from epub, epub to text, or ebook structure.
  extract from epub, epub to text, or ebook structure.
 license: MIT
-compatibility: >-
+compatibility: Python 3.8+ required. Core scripts use EbookLib (pip install EbookLib)
-  Python 3.8+ required. Core scripts use EbookLib (pip install EbookLib) for reading and
+  for reading and creating EPUBs. Optional epublib (pip install epublib) for non-intrusive
-  creating EPUBs. Optional epublib (pip install epublib) for non-intrusive editing.
+  editing. beautifulsoup4 (pip install beautifulsoup4) for text extraction. Optional
-  beautifulsoup4 (pip install beautifulsoup4) for text extraction. Optional LLM mode via
+  LLM mode via EPUB_LLM_URL + EPUB_LLM_KEY env vars (any OpenAI-compatible provider).
-  EPUB_LLM_URL + EPUB_LLM_KEY env vars (any OpenAI-compatible provider). EPUBCheck
+  EPUBCheck (Java, optional) for authoritative validation. Portable across all AgentSkills-compatible
  (Java, optional) for authoritative validation. Portable across all AgentSkills-compatible
  harnesses — scripts are platform-agnostic.
 metadata:
-  skills: [epub, ebook, publishing, knowledge-extraction, document-processing]
+  skills: epub, ebook, publishing, knowledge-extraction, document-processing
-  tags: [epub, ebook, epub3, epub2, publishing, document-extraction, knowledge]
+  tags: epub, ebook, epub3, epub2, publishing, document-extraction, knowledge
 ---
 # EPUB — Ebook Creation, Extraction & Enrichment
@@ -1,11 +1,13 @@
 ---
 name: forgejo-cli
-description: Use when managing a Forgejo or Gitea server from the terminal: issues, pull requests, repositories, file contents, labels, milestones, releases, webhooks, user settings, or any /api/v1 endpoint through a safe generic API command.
+description: 'Use when managing a Forgejo or Gitea server from the terminal: issues,
  pull requests, repositories, file contents, labels, milestones, releases, webhooks,
  user settings, or any /api/v1 endpoint through a safe generic API command.'
 license: MIT
 compatibility: Python 3.8+; requests is required only for live API calls.
 metadata:
  version: 2.0.0
-  tags: [forgejo, gitea, git, api, code-review]
+  tags: forgejo, gitea, git, api, code-review
 ---
 # Forgejo CLI v2
@@ -1,19 +1,15 @@
 ---
 name: ghost-cli
-description: >-
+description: Manage Ghost CMS content from the terminal — create and list posts, pages,
-  Manage Ghost CMS content from the terminal — create and list posts, pages,
+  and tags, and fetch site info via the Ghost Admin API (v5/v6). Use when the user
-  and tags, and fetch site info via the Ghost Admin API (v5/v6). Use when the
+  asks about ghost, cms, blog, blogging, posts, pages, tags, publishing, or site configuration.
  user asks about ghost, cms, blog, blogging, posts, pages, tags, publishing,
  or site configuration.
 license: MIT
-compatibility: Requires GHOST_URL and GHOST_ADMIN_KEY env vars. Admin key in
+compatibility: Requires GHOST_URL and GHOST_ADMIN_KEY env vars. Admin key in "id:secret"
-  "id:secret" format from Ghost Admin → Integrations. Python 3.8+ and the
+  format from Ghost Admin → Integrations. Python 3.8+ and the `requests` library.
  `requests` library.
 metadata:
-  tags: [ghost, cms, blog, blogging, post, page, tag, ghost-cms, content-management, api-client]
+  tags: ghost, cms, blog, blogging, post, page, tag, ghost-cms, content-management,
-  sources:
+    api-client
-    - https://ghost.org/docs/admin-api/
+  sources: https://ghost.org/docs/admin-api/, https://ghost.org/docs/
    - https://ghost.org/docs/
 ---
 # ghost-cli — Ghost CMS from the Terminal
@@ -27,3 +27,8 @@ Load this when setting up CI, troubleshooting runner registration failures, desi
 ## Requirements
 Linux, macOS, or Windows target hosts. Docker for containerized runners. Kubernetes for ARC deployments.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,16 +1,16 @@
 ---
 name: github-runner
-version: 1.0.2
+description: Deploy, manage, and troubleshoot self-hosted GitHub Actions runners.
-description: >-
+  Covers systemd service, Docker containers, Kubernetes (Actions Runner Controller),
-  Deploy, manage, and troubleshoot self-hosted GitHub Actions runners. Covers
+  and the Scale Set Client. Use when setting up a CI runner, debugging registration
-  systemd service, Docker containers, Kubernetes (Actions Runner Controller),
+  failures, designing autoscaling, or hardening runner security.
  and the Scale Set Client. Use when setting up a CI runner, debugging
  registration failures, designing autoscaling, or hardening runner security.
 license: MIT
-compatibility: Linux, macOS, or Windows target hosts. Docker for containerized runners. Kubernetes for ARC deployments.
+compatibility: Linux, macOS, or Windows target hosts. Docker for containerized runners.
  Kubernetes for ARC deployments.
 metadata:
-  tags: [github-actions, ci-cd, runners, devops, docker, kubernetes, autoscaling]
+  tags: github-actions, ci-cd, runners, devops, docker, kubernetes, autoscaling
  source: https://docs.github.com/en/actions/hosting-your-own-runners/managing-self-hosted-runners
  version: 1.0.2
 ---
 # Self-Hosted GitHub Actions Runner
@@ -1,26 +1,23 @@
 ---
 name: gutenberg
-description: >-
+description: Search, download, and extract public-domain books from Project Gutenberg.
-  Search, download, and extract public-domain books from Project Gutenberg.
+  Look up books by ID or keyword via gutendex, download plain-text and EPUB editions,
-  Look up books by ID or keyword via gutendex, download plain-text and EPUB
+  strip licensing boilerplate, extract clean text from EPUB for illustrated works,
-  editions, strip licensing boilerplate, extract clean text from EPUB for
+  and classify fiction vs non-fiction. Ships a portable CLI script with zero external
-  illustrated works, and classify fiction vs non-fiction. Ships a portable
+  dependencies. Use when the user says "gutenberg", "public domain", "download a book",
-  CLI script with zero external dependencies. Use when the user says
+  "classic literature", "free ebook", "gutenberg.org", or names any public-domain
-  "gutenberg", "public domain", "download a book", "classic literature",
+  title or author.
  "free ebook", "gutenberg.org", or names any public-domain title or author.
 license: MIT
-compatibility: >-
+compatibility: Python 3.8+ with zero external dependencies. The CLI uses only the
-  Python 3.8+ with zero external dependencies. The CLI uses only the Python
+  Python standard library (urllib.request, json, html.parser, zipfile, re, sys). For
-  standard library (urllib.request, json, html.parser, zipfile, re, sys).
+  EPUB extraction, Python 3.8+ with only stdlib is required (zipfile + html.parser).
-  For EPUB extraction, Python 3.8+ with only stdlib is required (zipfile +
+  The gutendex API (https://gutendex.com) requires no API key or registration. No
-  html.parser). The gutendex API (https://gutendex.com) requires no API key
+  env vars needed for basic operation.
  or registration. No env vars needed for basic operation.
 metadata:
-  tags: [gutenberg, project-gutenberg, books, public-domain, literature, classics, ebooks, text-extraction, epub]
+  tags: gutenberg, project-gutenberg, books, public-domain, literature, classics,
-  sources:
+    ebooks, text-extraction, epub
-    - https://gutendex.com
+  sources: https://gutendex.com, https://www.gutenberg.org
-    - https://www.gutenberg.org
+  skills: books, public-domain, literature, text-mining, ebooks
  skills: [books, public-domain, literature, text-mining, ebooks]
 ---
 # Gutenberg — Public Domain Book Toolkit
@@ -27,3 +27,13 @@ Haystack uses explicit Pipeline DAGs (add_component + connect) — different fro
 ## Requirements
 Python 3.8+ with `haystack-ai` package.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
@@ -28,3 +28,8 @@ Load this when working on a Hugo theme or site template layer.
 ## Requirements
 Hugo v0.154+. Works with any agent framework supporting the Agent Skills format.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,20 +1,17 @@
 ---
 name: jellyfin-cli
-description: >-
+description: Query your Jellyfin media server from the terminal — recently added movies
-  Query your Jellyfin media server from the terminal — recently added movies
+  and episodes, search across your library, browse libraries, and check server info
-  and episodes, search across your library, browse libraries, and check server
+  and stats. Use when the user asks about Jellyfin, media server, movies, TV shows,
-  info and stats. Use when the user asks about Jellyfin, media server, movies,
+  recently added content, or their media library.
  TV shows, recently added content, or their media library.
 license: MIT
-compatibility: Requires JELLYFIN_URL (default http://localhost:8096) and
+compatibility: Requires JELLYFIN_URL (default http://localhost:8096) and JELLYFIN_API_KEY
-  JELLYFIN_API_KEY env vars; Python 3.8+ and the `requests` library. Generate
+  env vars; Python 3.8+ and the `requests` library. Generate an API key at Dashboard
-  an API key at Dashboard → API Keys in the Jellyfin admin panel.
+  → API Keys in the Jellyfin admin panel.
 metadata:
-  tags: [jellyfin, media-server, movies, tv, episodes, recently-added, library,
+  tags: jellyfin, media-server, movies, tv, episodes, recently-added, library, home-media,
-    home-media, api-client]
+    api-client
-  sources:
+  sources: https://jellyfin.org/docs/general/clients/api, https://jellyfin.org/downloads
    - https://jellyfin.org/docs/general/clients/api
    - https://jellyfin.org/downloads
 ---
 # jellyfin-cli — Jellyfin Media Server from the Terminal
@@ -1,19 +1,16 @@
 ---
 name: jira-cli
-description: >-
+description: 'Interact with Atlassian Jira from the terminal: search issues, view
-  Interact with Atlassian Jira from the terminal: search issues, view
+  details, create issues, add comments, list projects, and transition status. Use
-  details, create issues, add comments, list projects, and transition
+  when the user mentions Jira, a ticket key (e.g. PROJ-123), or asks about issues,
-  status. Use when the user mentions Jira, a ticket key (e.g. PROJ-123),
+  bugs, tasks, projects, or sprint work.'
  or asks about issues, bugs, tasks, projects, or sprint work.
 license: MIT
-compatibility: Requires JIRA_EMAIL and JIRA_API_TOKEN env vars (free from
+compatibility: Requires JIRA_EMAIL and JIRA_API_TOKEN env vars (free from id.atlassian.com/manage/api-tokens),
-  id.atlassian.com/manage/api-tokens), Python 3.8+, and the `requests`
+  Python 3.8+, and the `requests` library. Also requires JIRA_SERVER (defaults to
-  library. Also requires JIRA_SERVER (defaults to your-domain.atlassian.net).
+  your-domain.atlassian.net).
 metadata:
-  tags: [jira, atlassian, issue-tracking, project-management, api-client]
+  tags: jira, atlassian, issue-tracking, project-management, api-client
-  sources:
+  sources: https://developer.atlassian.com/cloud/jira/platform/rest/v3/, https://id.atlassian.com/manage/api-tokens
    - https://developer.atlassian.com/cloud/jira/platform/rest/v3/
    - https://id.atlassian.com/manage/api-tokens
 ---
 # jira-cli — Jira Issue Tracker from the Terminal
@@ -27,3 +27,8 @@ Load this when writing, debugging, or optimizing JQL queries; building saved fil
 ## Requirements
 None — this is a reference skill. No scripts, no API keys.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -27,3 +27,8 @@ Load this when your team is struggling with throughput, cycle times are unpredic
 ## Requirements
 No special system requirements. Designed for agentic AI assistants.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -26,3 +26,13 @@ LangChain is the broadest LLM framework with 1000+ integrations. Its agents now
 ## Requirements
 Python 3.8+ with `langchain`, `langchain-community`, and provider-specific packages.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
@@ -29,3 +29,8 @@ Load this when designing agent architectures that need cycles, conditional branc
 ## Requirements
 Python 3.8+ with `langgraph`, `langchain`, and `langchain-openai` packages.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,17 +1,16 @@
 ---
 name: langgraph
-description: >-
+description: Build multi-agent AI systems with LangGraph — the low-level orchestration
-  Build multi-agent AI systems with LangGraph — the low-level orchestration
+  framework for stateful, graph-based agent workflows. Covers supervisor, swarm, and
-  framework for stateful, graph-based agent workflows. Covers supervisor, swarm,
+  hierarchical multi-agent patterns; subgraph composition; state management (checkpointers/stores);
-  and hierarchical multi-agent patterns; subgraph composition; state management
+  persistence; evals; and production debugging. Reach for this when designing agent
-  (checkpointers/stores); persistence; evals; and production debugging. Reach
+  architectures that need cycles, conditional branching, parallel execution, or human-in-the-loop
-  for this when designing agent architectures that need cycles, conditional
+  patterns.
  branching, parallel execution, or human-in-the-loop patterns.
 license: MIT
 version: "1.0.2"
 metadata:
  source: LangGraph by LangChain Inc — https://langchain.com/langgraph
-  spec-version: "1.0"
+  spec-version: '1.0'
  version: 1.0.2
 ---
 # LangGraph
@@ -1,19 +1,17 @@
 ---
 name: lastfm
-description: >-
+description: 'Interact with the Last.fm music data API: lookup user listening history,
-  Interact with the Last.fm music data API: lookup user listening history,
+  get artist/album/track metadata, discover similar music via collaborative filtering,
-  get artist/album/track metadata, discover similar music via collaborative
+  explore global and per-country charts, search by artist/album/track, manage tags,
-  filtering, explore global and per-country charts, search by artist/album/track,
+  and scrobble listening events. Use when the user asks about music data, listening
-  manage tags, and scrobble listening events. Use when the user asks about
+  statistics, music recommendations, similar artists, charts, or wants to scrobble
-  music data, listening statistics, music recommendations, similar artists,
+  or love tracks.'
  charts, or wants to scrobble or love tracks.
 license: MIT
-compatibility: Requires a Last.fm API key in the LASTFM_API_KEY env var.
+compatibility: Requires a Last.fm API key in the LASTFM_API_KEY env var. For write
-  For write operations (scrobble, love, now-playing), also needs
+  operations (scrobble, love, now-playing), also needs LASTFM_API_SECRET and LASTFM_SESSION_KEY.
-  LASTFM_API_SECRET and LASTFM_SESSION_KEY. The `lastfm-cli` CLI tool
+  The `lastfm-cli` CLI tool must be on PATH.
  must be on PATH.
 metadata:
-  tags: [music, lastfm, scrobbling, api-client, music-discovery]
+  tags: music, lastfm, scrobbling, api-client, music-discovery
  source: https://www.last.fm/api
 ---
@@ -0,0 +1,80 @@
 #!/usr/bin/env python3
 """Small, dependency-free Last.fm API client used by the lastfm skill."""
 import argparse
 import json
 import os
 import sys
 import urllib.parse
 import urllib.request
 API = "https://ws.audioscrobbler.com/2.0/"
 METHODS = {
    "user info": "user.getinfo", "user recent-tracks": "user.getrecenttracks",
    "user top-artists": "user.gettopartists", "user top-tracks": "user.gettoptracks",
    "user loved-tracks": "user.getlovedtracks", "user friends": "user.getfriends",
    "user weekly-charts": "user.getweeklychartlist", "artist similar": "artist.getsimilar",
    "artist info": "artist.getinfo", "artist top-tracks": "artist.gettoptracks",
    "artist search": "artist.search", "album info": "album.getinfo", "album search": "album.search",
    "track similar": "track.getsimilar", "track search": "track.search",
    "tag top-artists": "tag.gettopartists", "tag top-tracks": "tag.gettoptracks",
    "tag top-albums": "tag.gettopalbums", "chart top-artists": "chart.gettopartists",
    "geo top-artists": "geo.gettopartists", "geo top-tracks": "geo.gettoptracks",
 }
 def parser():
    p = argparse.ArgumentParser(description="Last.fm API client", epilog="Examples: lastfm-cli artist similar Radiohead --limit 5")
    p.add_argument("--json", action="store_true", help="emit JSON")
    p.add_argument("--dry-run", action="store_true", help="show the API plan without a request")
    subs = p.add_subparsers(dest="group", required=True)
    for group, commands in {"user": ["info", "recent-tracks", "top-artists", "top-tracks", "loved-tracks", "friends", "weekly-charts"], "artist": ["similar", "info", "top-tracks", "search"], "album": ["info", "search"], "track": ["similar", "search", "scrobble", "now-playing", "love"], "tag": ["top-artists", "top-tracks", "top-albums"], "chart": ["top-artists"], "geo": ["top-artists", "top-tracks"], "auth": ["get-token", "get-session"]}.items():
        gp = subs.add_parser(group)
        cs = gp.add_subparsers(dest="command", required=True)
        for command in commands:
            cp = cs.add_parser(command)
            if group in {"user", "artist", "album", "tag", "geo"} or group == "track":
                cp.add_argument("values", nargs="*", metavar="VALUE", help="username, artist, tag, country, or track values")
            cp.add_argument("--limit", type=int)
            cp.add_argument("--period")
            cp.add_argument("--autocorrect", action="store_true")
    return p
 def main():
    args = parser().parse_args()
    key = f"{args.group} {args.command}"
    if args.group == "auth":
        print("Auth operations require LASTFM_API_KEY, LASTFM_API_SECRET, and a browser authorization flow.")
        return 0
    values = getattr(args, "values", [])
    if args.command not in {"search", "top-artists", "top-tracks", "top-albums", "weekly-charts"} and not values:
        parser().error(f"{args.group} {args.command}: a value is required")
    method = METHODS.get(key)
    if not method:
        parser().error(f"unsupported operation: {key}")
    params = {"method": method, "format": "json"}
    if values:
        param = "user" if args.group == "user" else "country" if args.group == "geo" else "tag" if args.group == "tag" else "artist"
        params[param] = values[0]
        if args.group in {"track", "album"} and len(values) > 1:
            params["track" if args.group == "track" else "album"] = values[1]
    for option in ("limit", "period", "autocorrect"):
        value = getattr(args, option, None)
        if value not in (None, False): params[option] = value
    if args.dry_run:
        print(json.dumps({"dry_run": True, "url": API, "params": params}, indent=2))
        return 0
    api_key = os.getenv("LASTFM_API_KEY")
    if not api_key:
        print("LASTFM_API_KEY is required", file=sys.stderr)
        return 2
    params["api_key"] = api_key
    try:
        with urllib.request.urlopen(f"{API}?{urllib.parse.urlencode(params)}", timeout=30) as response:
            payload = json.load(response)
    except Exception as exc:
        print(f"Last.fm request failed: {exc}", file=sys.stderr)
        return 1
    print(json.dumps(payload, indent=2) if args.json else json.dumps(payload, indent=2))
    return 0
 if __name__ == "__main__":
    raise SystemExit(main())
@@ -31,8 +31,9 @@ check_output() {
    fi
 }
-# CLI path — adjust if not on PATH
+# Use the bundled CLI so the test is location-independent.
-CLI=lastfm-cli
+SKILL_DIR="$(cd "$(dirname "$0")/.." && pwd)"
 CLI="$SKILL_DIR/scripts/lastfm-cli"
 echo "=== lastfm-cli Test Suite ===
 "
@@ -68,4 +69,4 @@ check_output "Missing API key → graceful error" "LASTFM_API_KEY" $CLI user inf
 echo ""
 echo "=== Results: $PASS passed, $FAIL failed ==="
-[[ $FAIL -eq 0 ]] || exit 1
+[[ $FAIL -eq 0 ]] || exit 1
@@ -1,19 +1,16 @@
 ---
 name: lidarr-cli
-description: >-
+description: Manage your Lidarr music library from the terminal. Search and browse
-  Manage your Lidarr music library from the terminal. Search and browse artists
+  artists and albums, add new artists, check calendars, view queue and download history,
-  and albums, add new artists, check calendars, view queue and download history,
+  inspect quality and metadata profiles, and trigger searches. Use when the user mentions
-  inspect quality and metadata profiles, and trigger searches. Use when the user
+  Lidarr, music management, adding an artist, searching for music, checking the music
-  mentions Lidarr, music management, adding an artist, searching for music,
+  library, upcoming album releases, or download queue.
  checking the music library, upcoming album releases, or download queue.
 license: MIT
-compatibility: >-
+compatibility: Python 3.8+ with `requests` library. Requires ARR_SERVER_LIDARR and
  Python 3.8+ with `requests` library. Requires ARR_SERVER_LIDARR and
  ARR_KEY_LIDARR (from Lidarr → Settings → General → API Key).
 metadata:
-  tags: [lidarr, music, arr-stack, media-server, music-automation, api-client]
+  tags: lidarr, music, arr-stack, media-server, music-automation, api-client
-  sources:
+  sources: https://lidarr.audio/
    - https://lidarr.audio/
 ---
 # lidarr-cli — Lidarr Music Library Management
@@ -27,3 +27,13 @@ LlamaIndex evolved from a RAG indexing library into a full workflow framework. I
 ## Requirements
 Python 3.8+ with `llama_index` package.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
 ## Triggers
 Use this skill for the task types and keywords described in its SKILL.md description.
@@ -26,3 +26,8 @@ Load this when creating visuals in the Nous Research ecosystem — blog headers,
 ## Requirements
 Image generation backend for prompt-based workflows. Reference-image (img2img) workflows require an API supporting image inputs.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,27 +1,24 @@
 ---
 name: nous-branding
-description: >-
+description: 'Generate images and content consistent with the Nous Research brand
-  Generate images and content consistent with the Nous Research brand identity.
+  identity. Use when creating visuals in the Nous / Theia / Hermes ecosystem: a "cyber-classical"
  Use when creating visuals in the Nous / Theia / Hermes ecosystem: a "cyber-classical"
  style blending neo-classical statuary, cyberpunk/industrial grunge, and retro anime
  illustration. Covers official brand color palette, typography (Inter/IBM Plex Sans,
  JetBrains Mono, heavy distressed display faces), the Nous Girl mascot, texture system,
  and image prompt construction. Ships reference images for palette, mascot, and brand
-  collage that can be used as img2img inputs.
+  collage that can be used as img2img inputs.'
 license: MIT
-version: 1.1.0
+compatibility: Compatible with any agent capable of image generation or brand analysis.
-compatibility: Compatible with any agent capable of image generation or brand
+  Reference-image workflows (img2img, style transfer, variations) require an API endpoint
-  analysis. Reference-image workflows (img2img, style transfer, variations) require
+  supporting image inputs — use the assets/ images as input.
  an API endpoint supporting image inputs — use the assets/ images as input.
 metadata:
-  tags: [nous-research, theia, hermes, brand, illustration, mascot, image-generation, style-guide, cyber-classical]
+  tags: nous-research, theia, hermes, brand, illustration, mascot, image-generation,
-  sources:
+    style-guide, cyber-classical
-    - https://nousresearch.com
+  sources: https://nousresearch.com, {"Reference image"=>"assets/palette-typography-reference.png"},
-    - Reference image: assets/palette-typography-reference.png
+    {"Reference image"=>"assets/nous-girl-official.webp (official mascot, 2669×2709)"},
-    - Reference image: assets/nous-girl-official.webp (official mascot, 2669×2709)
+    {"Reference image"=>"assets/nous-girl-style-reference.png"}, {"Reference image"=>"assets/brand-collage-reference.png"},
-    - Reference image: assets/nous-girl-style-reference.png
+    https://nousresearch.com/wp-content/uploads/2024/03/NOUS-BRAND-BOOKLET-firstedition_1.pdf
-    - Reference image: assets/brand-collage-reference.png
+  version: 1.1.0
    - https://nousresearch.com/wp-content/uploads/2024/03/NOUS-BRAND-BOOKLET-firstedition_1.pdf
 ---
 # Nous Branding
@@ -28,3 +28,8 @@ Load this when working with OKF, creating knowledge bundles for AI agents, or co
 ## Requirements
 Python 3.8+ with PyYAML for validation. No API keys needed.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,21 +1,18 @@
 ---
 name: open-knowledge-format
-description: >-
+description: Google's Open Knowledge Format (OKF) v0.1 — an open, vendor-neutral spec
-  Google's Open Knowledge Format (OKF) v0.1 — an open, vendor-neutral spec for
+  for representing knowledge as markdown files with YAML frontmatter, designed for
-  representing knowledge as markdown files with YAML frontmatter, designed for
+  AI agent consumption. Use when the user mentions OKF, Open Knowledge Format, Google's
-  AI agent consumption. Use when the user mentions OKF, Open Knowledge Format,
+  knowledge format, LLM wiki bundles, agent knowledge packs, creating OKF bundles,
-  Google's knowledge format, LLM wiki bundles, agent knowledge packs, creating
+  validating OKF documents, or converting knowledge into the OKF standard.
  OKF bundles, validating OKF documents, or converting knowledge into the OKF
  standard.
 license: MIT
-compatibility: >-
+compatibility: Portable across any AgentSkills-compatible harness. The validation
-  Portable across any AgentSkills-compatible harness. The validation script
+  script requires Python 3.8+ with PyYAML (pip install PyYAML). All examples use standard
-  requires Python 3.8+ with PyYAML (pip install PyYAML). All examples use
+  markdown — readable in any editor or terminal.
  standard markdown — readable in any editor or terminal.
 metadata:
-  spec-version: "1.0"
+  spec-version: '1.0'
-  skills: [okf, open-knowledge-format, knowledge-format, agent-knowledge, llm-wiki]
+  skills: okf, open-knowledge-format, knowledge-format, agent-knowledge, llm-wiki
-  tags: [okf, knowledge-format, google, ai-agents, markdown, knowledge-management]
+  tags: okf, knowledge-format, google, ai-agents, markdown, knowledge-management
 ---
 # Open Knowledge Format (OKF) — v0.1
@@ -151,7 +148,7 @@ EOF
 cat > index.md << 'EOF'
 # Knowledge Bundle
-* [Sales Dataset](datasets/sales.md) - Retail sales data
+* Sales Dataset (`datasets/sales.md`) - Retail sales data in an example OKF bundle
 EOF
 # 4. Validate your bundle
@@ -1,18 +1,15 @@
 ---
 name: openlibrary-cli
-description: >-
+description: Search books, authors, and works on Open Library from the terminal. Look
-  Search books, authors, and works on Open Library from the terminal. Look up
+  up books by ISBN, search titles and authors, and fetch detailed work/author records
-  books by ISBN, search titles and authors, and fetch detailed work/author
+  via the public Open Library API. No API key required.
  records via the public Open Library API. No API key required.
 license: MIT
 compatibility: Python 3.8+ and the `requests` library. No API key or registration
-  required — the Open Library API is fully public. Optional OL_EMAIL env var
+  required — the Open Library API is fully public. Optional OL_EMAIL env var sets
-  sets a User-Agent contact for improved rate limiting.
+  a User-Agent contact for improved rate limiting.
 metadata:
-  tags: [open-library, books, authors, isbn, library, book-search, api-client, catalog]
+  tags: open-library, books, authors, isbn, library, book-search, api-client, catalog
-  sources:
+  sources: https://openlibrary.org/developers/api, https://openlibrary.org
    - https://openlibrary.org/developers/api
    - https://openlibrary.org
 ---
 # openlibrary-cli — Book Metadata from Open Library
@@ -27,3 +27,8 @@ Load this when filing an issue, opening a PR, or contributing to any open source
 ## Requirements
 None. Works with any agent supporting the Agent Skills format.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,18 +1,17 @@
 ---
 name: opensource-contributions
-description: >-
+description: Make good open source contributions — check CONTRIBUTING.md first, follow
-  Make good open source contributions — check CONTRIBUTING.md first, follow
+  project norms, be a good citizen. Covers bug reports, feature requests, and pull
-  project norms, be a good citizen. Covers bug reports, feature requests, and
+  requests with a defensible default posture when the project hasn't documented expectations.
  pull requests with a defensible default posture when the project hasn't
  documented expectations.
 license: MIT
-compatibility: Compatible with any agent supporting the Agent Skills format (Hermes Agent, Claude Code, GitHub Copilot, OpenCode, Cursor, etc.)
+compatibility: Compatible with any agent supporting the Agent Skills format (Hermes
  Agent, Claude Code, GitHub Copilot, OpenCode, Cursor, etc.)
 metadata:
-  tags: [opensource, contributing, github, etiquette, PRs, issues]
+  tags: opensource, contributing, github, etiquette, PRs, issues
-  related_skills: [github-issues, github-pr-workflow, github-code-review]
+  related_skills: github-issues, github-pr-workflow, github-code-review
-  version: "1.6.0"
+  version: 1.6.0
  author: agent-skills
-source_repo: https://agentskills.io
+  source_repo: https://agentskills.io
 ---
 # Open Source Contributions
@@ -1,20 +1,15 @@
 ---
 name: peertube
-description: >-
+description: 'Browse PeerTube federated video from the terminal: view videos and channels,
-  Browse PeerTube federated video from the terminal: view videos and
+  search across instances, check server stats, and manage your account. Uses OAuth2
-  channels, search across instances, check server stats, and manage your
+  authentication with token persistence. Use when the user mentions PeerTube, federated
-  account. Uses OAuth2 authentication with token persistence. Use when the
+  video, decentralized video platforms, or browsing/uploading to a PeerTube instance.'
  user mentions PeerTube, federated video, decentralized video platforms,
  or browsing/uploading to a PeerTube instance.
 license: MIT
-compatibility: Requires PEERTUBE_SERVER env var (set to your instance URL,
+compatibility: Requires PEERTUBE_SERVER env var (set to your instance URL, e.g. https://watch.nousresearch.com),
-  e.g. https://watch.nousresearch.com), Python 3.8+, and the `requests`
+  Python 3.8+, and the `requests` library. OAuth2 tokens persisted to ~/.config/peertube-cli/token.json.
  library. OAuth2 tokens persisted to ~/.config/peertube-cli/token.json.
 metadata:
-  tags: [peertube, federated-video, video-platform, activitypub, api-client]
+  tags: peertube, federated-video, video-platform, activitypub, api-client
-  sources:
+  sources: https://joinpeertube.org/, https://docs.joinpeertube.org/api/reference
    - https://joinpeertube.org/
    - https://docs.joinpeertube.org/api/reference
 ---
 # peertube-cli — PeerTube Federated Video
@@ -28,3 +28,8 @@ Load this when starting product discovery, needing to interview stakeholders, or
 ## Requirements
 None. Agent-agnostic — works with any spec-driven or requirements pipeline.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,19 +1,16 @@
 ---
 name: prowlarr-cli
-description: >-
+description: Manage your Prowlarr indexer hub from the terminal. List and inspect
-  Manage your Prowlarr indexer hub from the terminal. List and inspect indexers,
+  indexers, view query/grab statistics, check health, manage connected *arr applications
-  view query/grab statistics, check health, manage connected *arr applications
+  and download clients, browse search history, and test indexer connectivity. Use
-  and download clients, browse search history, and test indexer connectivity.
+  when the user mentions Prowlarr, indexers, indexer management, testing indexers,
-  Use when the user mentions Prowlarr, indexers, indexer management, testing
+  search history, or the indexer hub.
  indexers, search history, or the indexer hub.
 license: MIT
-compatibility: >-
+compatibility: Python 3.8+ with `requests` library. Requires ARR_SERVER_PROWLARR and
  Python 3.8+ with `requests` library. Requires ARR_SERVER_PROWLARR and
  ARR_KEY_PROWLARR (from Prowlarr → Settings → General → API Key).
 metadata:
-  tags: [prowlarr, indexer, arr-stack, media-server, api-client]
+  tags: prowlarr, indexer, arr-stack, media-server, api-client
-  sources:
+  sources: https://prowlarr.com/
    - https://prowlarr.com/
 ---
 # prowlarr-cli — Prowlarr Indexer Management
@@ -29,3 +29,8 @@ Load this whenever building agents, tool-using LLM workflows, or graph-based sta
 ## Requirements
 Python 3.10+ with `pydantic-ai` or `pydantic-ai-slim` package.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -0,0 +1,70 @@
 #!/usr/bin/env ruby
 # frozen_string_literal: true
 require "yaml"
 ROOT = File.expand_path("..", __dir__)
 ALLOWED_FIELDS = %w[name description license compatibility metadata allowed-tools].freeze
 README_HEADINGS = ["Why Install This Skill", "What You Get", "Quick Start", "Triggers", "Requirements"].freeze
 errors = []
 Dir.glob("#{ROOT}/**/SKILL.md").sort.each do |skill|
  relative = skill.delete_prefix("#{ROOT}/")
  text = File.read(skill)
  match = text.match(/\A---\n(.*?)\n---\n/m)
  unless match
    errors << "#{relative}: missing YAML frontmatter"
    next
  end
  begin
    data = YAML.safe_load(match[1], permitted_classes: [], aliases: false)
  rescue Psych::Exception => e
    errors << "#{relative}: invalid YAML: #{e.message.lines.first.strip}"
    next
  end
  unless data.is_a?(Hash)
    errors << "#{relative}: frontmatter must be a mapping"
    next
  end
  name = data["name"]
  expected_name = File.basename(File.dirname(skill))
  errors << "#{relative}: invalid name" unless name.is_a?(String) && name.match?(/\A[a-z0-9]+(?:-[a-z0-9]+)*\z/) && name.length.between?(1, 64) && name == expected_name
  description = data["description"]
  errors << "#{relative}: invalid description" unless description.is_a?(String) && description.length.between?(1, 1024)
  errors << "#{relative}: unsupported field(s): #{(data.keys - ALLOWED_FIELDS).join(', ')}" unless (data.keys - ALLOWED_FIELDS).empty?
  if data.key?("compatibility") && (!data["compatibility"].is_a?(String) || !data["compatibility"].length.between?(1, 500))
    errors << "#{relative}: invalid compatibility"
  end
  if data.key?("metadata") && (!data["metadata"].is_a?(Hash) || !data["metadata"].all? { |key, value| key.is_a?(String) && value.is_a?(String) })
    errors << "#{relative}: metadata must map strings to strings"
  end
  errors << "#{relative}: must stay under 500 lines" unless text.lines.length < 500
  root = File.dirname(skill)
  text.scan(/\[[^\]]+\]\(([^)]+)\)/).flatten.each do |target|
    path = target.split("#", 2).first
    next if path.empty? || path.match?(%r{\A(?:https?:|mailto:|/)})
    errors << "#{relative}: missing linked resource #{path}" unless File.exist?(File.expand_path(path, root))
  end
  readme = File.join(root, "README.md")
  unless File.file?(readme)
    errors << "#{relative}: missing README.md"
    next
  end
  readme_text = File.read(readme)
  README_HEADINGS.each do |heading|
    errors << "#{relative}: README missing #{heading}" unless readme_text.match?(/^#+ #{Regexp.escape(heading)}\s*$/)
  end
 end
 if errors.empty?
  puts "Validated #{Dir.glob("#{ROOT}/**/SKILL.md").length} skills."
 else
  warn errors.join("\n")
  exit 1
 end
@@ -27,3 +27,8 @@ Load this when you need to understand how a codebase works, produce a design doc
 ## Requirements
 Git, a programming language runtime matching the target codebase.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,17 +1,16 @@
 ---
 name: software-architecture-analysis
-description: >-
+description: Reverse-engineer a software codebase to understand architecture, data
-  Reverse-engineer a software codebase to understand architecture, data flow,
+  flow, privacy posture, and feature surface — then produce a clean-room design document,
-  privacy posture, and feature surface — then produce a clean-room design
+  PRD, or migration plan that re-imagines the system under new constraints (local-first,
-  document, PRD, or migration plan that re-imagines the system under new
+  privacy-first, self-hosted). Use when you need to understand how a system works
-  constraints (local-first, privacy-first, self-hosted). Use when you need
+  from its source code or produce a specification without copying implementation details.
  to understand how a system works from its source code or produce a
  specification without copying implementation details.
 license: MIT
-compatibility: Requires git, a programming language runtime matching the
+compatibility: Requires git, a programming language runtime matching the target codebase,
-  target codebase, and a markdown editor for output.
+  and a markdown editor for output.
 metadata:
-  tags: [reverse-engineering, architecture, prd, design-document, codebase-analysis, clean-room]
+  tags: reverse-engineering, architecture, prd, design-document, codebase-analysis,
    clean-room
 ---
 # Software Architecture Analysis — Codebase Reverse Engineering to Design Document
@@ -29,3 +29,8 @@ Load this when designing a spec-driven pipeline, writing an AI-ready specificati
 ## Requirements
 Tool-agnostic — works with Claude Code, Cursor, Hermes Agent, Devin, OpenHands, droid, or any AI coding agent.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,19 +1,19 @@
 ---
 name: spec-driven-development
-description: >-
+description: Spec-Driven Development (SDD) methodology for AI software factories —
-  Spec-Driven Development (SDD) methodology for AI software factories — where
+  where structured specifications are the input, AI agents generate the code, and
-  structured specifications are the input, AI agents generate the code, and
+  quality gates enforce correctness at each pipeline phase. Use when designing a spec
-  quality gates enforce correctness at each pipeline phase. Use when designing
+  → review → decompose → implement → verify pipeline that any AI coding tool (Claude
-  a spec → review → decompose → implement → verify pipeline that any AI
+  Code, Cursor, Hermes Agent, Devin, OpenHands) can follow.
  coding tool (Claude Code, Cursor, Hermes Agent, Devin, OpenHands) can follow.
 license: MIT
-compatibility: Tool-agnostic — methodology applies to any AI coding agent. Templates use markdown and Gherkin. Scripts require bash.
+compatibility: Tool-agnostic — methodology applies to any AI coding agent. Templates
  use markdown and Gherkin. Scripts require bash.
 metadata:
  source: https://github.com/magnus919/agent-skills/spec-driven-development
-  spec-version: "1.2.0"
+  spec-version: 1.2.0
-  hermes:
+  hermes: tags=["sdd", "specifications", "ai-code-generation", "software-factory",
-    tags: [sdd, specifications, ai-code-generation, software-factory, bdd, gherkin, quality-gates, phase-gates]
+    "bdd", "gherkin", "quality-gates", "phase-gates"], related_skills=["sdd-authoring",
-    related_skills: [sdd-authoring, sdd-review, sdd-verification, sdd-work-decomposition]
+    "sdd-review", "sdd-verification", "sdd-work-decomposition"]
 ---
 # Spec-Driven Development for AI Software Factories
@@ -26,3 +26,8 @@ Load this for ANY technical issue — test failures, production bugs, unexpected
 ## Requirements
 Platform-agnostic. Some sections cover macOS-specific sandbox debugging patterns. Requires access to source code and testing tools.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,26 +1,18 @@
 ---
 name: systematic-debugging
-description: >-
+description: '4-phase root cause debugging protocol: understand bugs before fixing.
-  4-phase root cause debugging protocol: understand bugs before fixing.
+  Use for ANY technical issue — test failures, production bugs, unexpected behavior,
-  Use for ANY technical issue — test failures, production bugs, unexpected
+  performance problems, build failures, or integration issues. ESPECIALLY when under
-  behavior, performance problems, build failures, or integration issues.
+  time pressure, when "one quick fix" seems obvious, or when previous fix attempts
-  ESPECIALLY when under time pressure, when "one quick fix" seems obvious,
+  have failed.'
  or when previous fix attempts have failed.
 license: MIT
-compatibility: >-
+compatibility: Platform-agnostic. Some sections cover macOS-specific sandbox debugging.
-  Platform-agnostic. Some sections cover macOS-specific sandbox debugging.
+  Requires access to source code, version control (git), and testing tools appropriate
-  Requires access to source code, version control (git), and testing tools
+  to the project.
  appropriate to the project.
 metadata:
-  tags:
+  tags: debugging, troubleshooting, problem-solving, root-cause, investigation
-    - debugging
+  source: Adapted from obra/superpowers (Jesse Vincent, MIT). Expanded with real-world
-    - troubleshooting
+    debugging patterns from production use.
    - problem-solving
    - root-cause
    - investigation
  source: >-
    Adapted from obra/superpowers (Jesse Vincent, MIT).
    Expanded with real-world debugging patterns from production use.
 ---
 # Systematic Debugging
@@ -1,19 +1,15 @@
 ---
 name: tempest-cli
-description: >-
+description: 'Query hyper-local weather from a WeatherFlow Tempest station: current
-  Query hyper-local weather from a WeatherFlow Tempest station: current
+  conditions, 7-day forecast, historical observations, and real-time UDP broadcasts.
-  conditions, 7-day forecast, historical observations, and real-time UDP
+  Use when the user asks about the weather, temperature, rain, wind, humidity, forecast,
-  broadcasts. Use when the user asks about the weather, temperature, rain,
+  or wants conditions from their own station rather than a generic weather service.'
  wind, humidity, forecast, or wants conditions from their own station
  rather than a generic weather service.
 license: MIT
-compatibility: Requires TEMPEST_TOKEN env var (free from weatherflow.com),
+compatibility: Requires TEMPEST_TOKEN env var (free from weatherflow.com), Python
-  Python 3.8+, and the `requests` library.
+  3.8+, and the `requests` library.
 metadata:
-  tags: [weather, tempest, forecast, weatherflow, station, hyper-local]
+  tags: weather, tempest, forecast, weatherflow, station, hyper-local
-  sources:
+  sources: https://weatherflow.com, https://swd.weatherflow.com/swd/rest
    - https://weatherflow.com
    - https://swd.weatherflow.com/swd/rest
 ---
 # tempest-cli — Hyper-Local Weather from Your Tempest Station
@@ -1,18 +1,14 @@
 ---
 name: tmdb-cli
-description: >-
+description: Search and discover movies, TV shows, and trending content via The Movie
-  Search and discover movies, TV shows, and trending content via The Movie
+  Database (TMDb) API v3. Use when the user asks about movies, TV, film, cinema, genres,
-  Database (TMDb) API v3. Use when the user asks about movies, TV, film,
+  certifications, ratings, cast, upcoming releases, or trending media.
  cinema, genres, certifications, ratings, cast, upcoming releases, or
  trending media.
 license: MIT
-compatibility: Requires TMDB_ACCESS_TOKEN or TMDB_API_KEY env var (free at
+compatibility: Requires TMDB_ACCESS_TOKEN or TMDB_API_KEY env var (free at themoviedb.org/settings/api),
-  themoviedb.org/settings/api), Python 3.8+, and the `requests` library.
+  Python 3.8+, and the `requests` library.
 metadata:
-  tags: [tmdb, movies, tv, film, cinema, entertainment, media-discovery, api-client]
+  tags: tmdb, movies, tv, film, cinema, entertainment, media-discovery, api-client
-  sources:
+  sources: https://developer.themoviedb.org/reference, https://www.themoviedb.org/settings/api
    - https://developer.themoviedb.org/reference
    - https://www.themoviedb.org/settings/api
 ---
 # tmdb-cli — Movie & TV Discovery from the Terminal
@@ -29,3 +29,8 @@ Load this when setting up or debugging a Traefik instance for reverse proxy, loa
 ## Requirements
 Docker for containerized deployment. Standard Linux server for native installation.
 ## Quick Start
 Start with the setup and first workflow in SKILL.md, then use the linked resources for the specific task you need to complete.
@@ -1,16 +1,14 @@
 ---
 name: traefik
-description: >-
+description: Deploy, configure, and troubleshoot Traefik v3 reverse proxy — covers
-  Deploy, configure, and troubleshoot Traefik v3 reverse proxy — covers all
+  all providers, routing, TLS/ACME, middlewares, and production patterns with YAML
-  providers, routing, TLS/ACME, middlewares, and production patterns with
+  examples. Load when setting up or debugging a Traefik instance.
  YAML examples. Load when setting up or debugging a Traefik instance.
 license: MIT
-version: 0.1.0
+compatibility: Compatible with any agent supporting the Agent Skills format (Hermes
-compatibility: >-
+  Agent, Claude Code, GitHub Copilot, OpenCode, Cursor, etc.)
  Compatible with any agent supporting the Agent Skills format (Hermes Agent,
  Claude Code, GitHub Copilot, OpenCode, Cursor, etc.)
 metadata:
  source: https://doc.traefik.io/traefik/
  version: 0.1.0
 ---
 # Traefik Agent Skill
@@ -1,19 +1,16 @@
 ---
 name: trakt
-description: >-
+description: Discover trending, anticipated, and popular movies and TV shows via the
-  Discover trending, anticipated, and popular movies and TV shows via the
+  Trakt.tv API from the terminal. No authentication required for read-only discovery.
-  Trakt.tv API from the terminal. No authentication required for read-only
+  Use when the user asks about what to watch, trending movies, popular shows, or media
-  discovery. Use when the user asks about what to watch, trending movies,
+  discovery.
  popular shows, or media discovery.
 license: MIT
-compatibility: Requires TRAKT_CLIENT_ID env var (free from
+compatibility: Requires TRAKT_CLIENT_ID env var (free from trakt.tv/oauth/applications),
-  trakt.tv/oauth/applications), Python 3.8+, and the `requests` library.
+  Python 3.8+, and the `requests` library. No OAuth or user login needed for discovery
-  No OAuth or user login needed for discovery endpoints.
+  endpoints.
 metadata:
-  tags: [trakt, media-discovery, movies, tv-shows, trending, api-client]
+  tags: trakt, media-discovery, movies, tv-shows, trending, api-client
-  sources:
+  sources: https://trakt.tv/, https://trakt.docs.apiary.io/
    - https://trakt.tv/
    - https://trakt.docs.apiary.io/
 ---
 # trakt-cli — Trakt.tv Media Discovery
@@ -1,19 +1,15 @@
 ---
 name: transistor
-description: >-
+description: 'Manage Transistor.fm podcast hosting from the terminal: view shows,
-  Manage Transistor.fm podcast hosting from the terminal: view shows, list
+  list episodes, check analytics, and get subscriber counts. Use when the user mentions
-  episodes, check analytics, and get subscriber counts. Use when the user
+  Transistor, podcast hosting, podcast analytics, show management, or episode tracking.'
  mentions Transistor, podcast hosting, podcast analytics, show management,
  or episode tracking.
 license: MIT
-compatibility: Requires TRANSISTOR_API_KEY env var (from Settings → API Keys
+compatibility: Requires TRANSISTOR_API_KEY env var (from Settings → API Keys in the
-  in the Transistor.fm dashboard), Python 3.8+, and the `requests` library.
+  Transistor.fm dashboard), Python 3.8+, and the `requests` library. Uses the Transistor.fm
-  Uses the Transistor.fm v1 REST API with JSON:API format responses.
+  v1 REST API with JSON:API format responses.
 metadata:
-  tags: [transistor, podcast, podcast-hosting, analytics, api-client]
+  tags: transistor, podcast, podcast-hosting, analytics, api-client
-  sources:
+  sources: https://transistor.fm/, https://developers.transistor.fm/
    - https://transistor.fm/
    - https://developers.transistor.fm/
 ---
 # transistor-cli — Transistor.fm Podcast Hosting
@@ -1,25 +1,21 @@
 ---
 name: yc-default-alive-calculator
-description: >-
+description: Evaluate whether a startup is on a trajectory to profitability before
-  Evaluate whether a startup is on a trajectory to profitability before running
+  running out of cash — Paul Graham's "Default Alive / Default Dead" framework. Takes
-  out of cash — Paul Graham's "Default Alive / Default Dead" framework. Takes
+  revenue, burn rate, cash on hand, and growth rate; computes runway, burn multiple,
-  revenue, burn rate, cash on hand, and growth rate; computes runway, burn
+  and months to breakeven. Ships a deterministic CLI calculator. Load when founders
-  multiple, and months to breakeven. Ships a deterministic CLI calculator.
+  ask about runway, burn rate, default alive, whether they need to raise money, or
-  Load when founders ask about runway, burn rate, default alive, whether they
+  financial sustainability analysis.
  need to raise money, or financial sustainability analysis.
 license: MIT
 compatibility: Python 3.9+ with standard library only (no external dependencies).
  The default-alive.py script uses only math, json, and sys.
 metadata:
-  spec-version: "1.0"
+  spec-version: '1.0'
-  tags: [startup-finance, runway-analysis, ycombinator, paul-graham, fundraising,
+  tags: startup-finance, runway-analysis, ycombinator, paul-graham, fundraising, financial-modeling,
-         financial-modeling, default-alive, burn-rate, startup-metrics]
+    default-alive, burn-rate, startup-metrics
-  sources:
+  sources: https://paulgraham.com/default.html, https://paulgraham.com/die.html, https://www.ycombinator.com/about
-    - https://paulgraham.com/default.html
+  skills: yc-weekly-growth-compass
-    - https://paulgraham.com/die.html
+  requires-toolsets: terminal
    - https://www.ycombinator.com/about
  skills: [yc-weekly-growth-compass]
  requires-toolsets: [terminal]
 ---
 # Default Alive / Default Dead Calculator
@@ -1,26 +1,21 @@
 ---
 name: yc-weekly-growth-compass
-description: >-
+description: Paul Graham's "Startup = Growth" framework as an operational tool. Computes
-  Paul Graham's "Startup = Growth" framework as an operational tool. Computes
+  weekly growth rates, benchmarks against YC tiers (1% concerning, 5-7% good, 10%+
-  weekly growth rates, benchmarks against YC tiers (1% concerning, 5-7% good,
+  exceptional), projects compound growth over time, and frames every startup decision
-  10%+ exceptional), projects compound growth over time, and frames every
+  through the question "does this serve your target growth rate?" Ships a deterministic
-  startup decision through the question "does this serve your target growth
+  CLI calculator. Load when founders ask about growth rate, weekly growth, startup
-  rate?" Ships a deterministic CLI calculator. Load when founders ask about
+  traction, metrics, or whether they're moving fast enough.
  growth rate, weekly growth, startup traction, metrics, or whether they're
  moving fast enough.
 license: MIT
 compatibility: Python 3.9+ with standard library only (no external dependencies).
  The growth-compass.py script uses only math, json, and sys.
 metadata:
-  spec-version: "1.0"
+  spec-version: '1.0'
-  tags: [startup-growth, growth-rate, ycombinator, paul-graham, startup-metrics,
+  tags: startup-growth, growth-rate, ycombinator, paul-graham, startup-metrics, weekly-growth,
-         weekly-growth, traction, compound-growth, startup-compass]
+    traction, compound-growth, startup-compass
-  sources:
+  sources: https://paulgraham.com/growth.html, https://paulgraham.com/ds.html, https://www.ycombinator.com/about
-    - https://paulgraham.com/growth.html
+  skills: yc-default-alive-calculator
-    - https://paulgraham.com/ds.html
+  requires-toolsets: terminal
    - https://www.ycombinator.com/about
  skills: [yc-default-alive-calculator]
  requires-toolsets: [terminal]
 ---
 # Weekly Growth Compass