skills/magnus919_agent-skills
1
0
Fork 0
mirror of https://github.com/magnus919/agent-skills.git synced 2026-09-11 19:47:12 +03:00
Code Issues Packages Projects Releases Wiki Activity
257 Commits 11 Branches 10 Tags
c3f6a30826c6bd27059cfee082430a6b258b06c5
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Magnus HedemarkandGitHub c3f6a30826 feat: add strategy-frameworks skill (#29) 
Closes #7
 2026-07-13 03:11:49 -04:00
Magnus HedemarkandGitHub d89bf34cb8 feat: add product-methodology skill (#27) 
Closes #6.\n\nImplemented with OpenCode and specialist research; corrected and binding-reviewed by Jasper.
 2026-07-13 02:23:05 -04:00
Magnus HedemarkandGitHub 3b0b743687 feat: add financial-modeling skill (#26) 
Closes #5.\n\nPorted and revised with researcher, OpenCode, Jasper self-review, and independent verifier assistance.
 2026-07-13 01:58:10 -04:00
Magnus HedemarkandGitHub 1624ef76f9 feat: add seo-audit skill (#19) 
* feat: add seo-audit skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove host-specific port assumptions

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: generalize seo examples

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: anonymize site-specific seo examples

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: generalize portable examples

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:46:41 -04:00
Magnus HedemarkandGitHub fd0b236745 feat: add verification-methodology skill (#18) 
* feat: add verification-methodology skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: harden portable skill guidance

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:46:16 -04:00
Magnus HedemarkandGitHub 6c19d55f43 feat: add technology-radar skill (#17) 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:45:52 -04:00
Magnus HedemarkandGitHub 8786ad1fd1 feat: add c4-diagramming skill (#16) 
* feat: add c4-diagramming skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove host-specific port assumptions

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:45:28 -04:00
Magnus HedemarkandGitHub b5ba3a7793 feat: add adr-authoring skill (#15) 
* feat: add adr-authoring skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove host-specific port assumptions

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:45:03 -04:00
Magnus HedemarkandGitHub 7057f3ce86 feat: add mermaid-diagrams skill (#14) 
* feat: add mermaid-diagrams skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: harden portable skill guidance

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove nonportable source assumptions

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: generalize portable examples

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:44:39 -04:00
Magnus HedemarkandGitHub e517f4c551 feat: add security-audit-methodology skill (#13) 
* feat: add security-audit-methodology skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: harden portable skill guidance

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:44:14 -04:00
Magnus HedemarkandGitHub 306b1bdbee feat: add technical-documentation skill (#12) 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:43:31 -04:00
Magnus HedemarkandGitHub e863aab851 feat: add research-methodology skill (#11) 
* feat: add research-methodology skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove nonportable source assumptions

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:37:23 -04:00
Magnus HedemarkandGitHub 6fd559e903 feat: add site-reliability-engineering skill (#10) 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:34:26 -04:00
Magnus HedemarkandGitHub b1d125caf3 feat: add artifact-pyramids skill (#9) 
* feat: add artifact-pyramids skill

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

* fix: remove host-specific artifact references

Signed-off-by: Magnus Hedemark <magnus919@pm.me>

---------

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-12 14:33:06 -04:00
Magnus HedemarkandGitHub 71002718e9 fix: SkillOpt Epoch 1 — harden meshcore packet capture skill (#4) 2026-07-12 00:36:01 -04:00
Magnus HedemarkandGitHub fcc40d71ca fix: route Kubernetes endpoint verification failures explicitly 
Held-out SkillOpt validation found and closed the endpoint-routing gap. Repository validation passed.
 2026-07-11 18:53:15 -04:00
Magnus HedemarkandGitHub c4971a7f12 Merge pull request #2 from magnus919/feat/skillopt-kubernetes 
fix: SkillOpt hygiene cleanup for Kubernetes skill
 2026-07-11 18:44:23 -04:00
Magnus Hedemark b3b98ed198 fix: SkillOpt hygiene cleanup for Kubernetes skill 2026-07-11 18:43:57 -04:00
Magnus HedemarkandGitHub 9b7bd4cfba Merge pull request #1 from magnus919/chore/github-community-setup 
docs: prepare repository for GitHub contributors
 2026-07-11 18:02:43 -04:00
Magnus Hedemark a25a690ca6 docs: link contributor guidance from README 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-11 18:00:03 -04:00
Magnus Hedemark df1b5272e4 docs: prepare repository for GitHub contributors 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-11 17:55:18 -04:00
Magnus Hedemark 61bb01b0ac fix(three): explain skill value 2026-07-11 17:22:38 -04:00
Magnus Hedemark 55fdf36417 fix(three): document skill contents 2026-07-11 17:22:38 -04:00
Magnus Hedemark 12926f31bf fix(three): document quick start 2026-07-11 17:22:38 -04:00
Magnus Hedemark 09ab025633 fix(three): add required quick start 2026-07-11 17:22:38 -04:00
Magnus Hedemark 69f043a2c5 feat(three): add browser 3D scene skill 2026-07-11 17:22:38 -04:00
Magnus Hedemark 1dd15688b4 feat(skills): add failure-mode routing index 2026-07-11 17:18:51 -04:00
Magnus Hedemark c8ab32f8f8 docs(skills): add exit criteria example 2026-07-11 17:11:34 -04:00
Magnus Hedemark 886f295894 docs(skills): add applicability example 2026-07-11 17:11:33 -04:00
Magnus Hedemark fc1f7ba389 docs(skills): keep confirmation scope focused 2026-07-11 17:11:31 -04:00
Magnus Hedemark 711505865f docs(skills): document applicability sections 2026-07-11 17:11:31 -04:00
Magnus Hedemark a49089ea0f docs(skills): complete confirmation field guidance 2026-07-11 17:11:15 -04:00
Magnus Hedemark 06f10c1192 docs(skills): specify confirmation frontmatter 2026-07-11 17:10:58 -04:00
Magnus Hedemark 15c5cf9f2c fix(workflows): explain bundle value 2026-07-11 17:09:36 -04:00
Magnus Hedemark 50b1fbfa9f fix(flaresolverr): explain skill value 2026-07-11 17:09:36 -04:00
Magnus Hedemark 275129c81e fix(workflows): document bundle contents 2026-07-11 17:09:13 -04:00
Magnus Hedemark 6d1e69da66 fix(flaresolverr): document skill contents 2026-07-11 17:09:12 -04:00
Magnus Hedemark 60730e45ad fix(workflows): document quick start 2026-07-11 17:08:07 -04:00
Magnus Hedemark 31641ab84a fix(flaresolverr): document quick start 2026-07-11 17:08:06 -04:00
Magnus Hedemark 61b38784cd fix(workflows): add bundle quick start 2026-07-11 17:07:37 -04:00
Magnus Hedemark 79ede71dff fix(flaresolverr): add required quick start 2026-07-11 17:07:36 -04:00
Magnus Hedemark 7f41005192 feat(workflows): add research-and-vault orchestrator 2026-07-11 17:03:05 -04:00
Magnus Hedemark f0efb25d91 docs(skills): define completion and exit conditions 2026-07-11 17:02:59 -04:00
Magnus Hedemark 87051aeba8 docs(skills): define use-when guidance 2026-07-11 17:02:57 -04:00
Magnus Hedemark 0546ddcb9d docs(skills): standardize confirmation gates 2026-07-11 17:02:54 -04:00
Magnus Hedemark 7704ded9c0 feat(flaresolverr): add browser challenge CLI skill 2026-07-11 17:01:03 -04:00
Magnus Hedemark 3497d9ff19 docs: deepen Kubernetes production safety guidance 2026-07-11 16:33:46 -04:00
Magnus Hedemark d6bc6ffe8b feat: add comprehensive Kubernetes agent skill 2026-07-11 16:27:12 -04:00
Magnus Hedemark bcfc61b54d fix(forgejo-cli): harden API transport and guidance 2026-07-11 13:24:50 -04:00
Magnus Hedemark a5f7e74972 fix: resolve skill audit findings 2026-07-11 09:24:58 -04:00
Magnus Hedemark 5dfea171cb docs: harden agent-skills reference 2026-07-11 09:07:10 -04:00
Magnus Hedemark e7387a22b4 fix: tighten Forgejo CLI request validation 2026-07-11 08:48:15 -04:00
Magnus Hedemark a77a890194 fix: complete Forgejo CLI API contracts 2026-07-11 08:44:13 -04:00
Magnus Hedemark 3f40fbb3be feat: rebuild forgejo CLI for v2 2026-07-11 08:35:48 -04:00
Magnus Hedemark 2d35f41be0 fix: tighten compose skill review findings 2026-07-11 08:14:13 -04:00
Magnus Hedemark 0094741ce8 fix: SkillOpt Epoch 3 — gate destructive compose cleanup 2026-07-11 07:58:31 -04:00
Magnus Hedemark 9d015a8cec fix: SkillOpt Epoch 2 — surface compose decisions 2026-07-11 07:58:31 -04:00
Magnus Hedemark 6c343bb455 fix: SkillOpt Epoch 1 — improve compose defaults 2026-07-11 07:58:31 -04:00
Magnus Hedemark 7b1a2c3c75 fix: keep compose watch template compatible 2026-07-11 07:58:31 -04:00
Magnus Hedemark f340dc1c62 feat: add portable docker-compose skill 2026-07-11 07:58:31 -04:00
Magnus Hedemark c4cd14b5b1 fix(skills): repair audited format and resource defects 2026-07-10 21:50:56 -04:00
Magnus Hedemark 1305c770d1 fix(raleigh): add JSON output for dataset info 2026-07-10 21:31:43 -04:00
Magnus Hedemark 12e506186b docs(hugo-theme): document cover images and template blocks 2026-07-10 21:25:42 -04:00
Magnus Hedemark 3109e4baa8 fix(workflow-architect): restore valid skill markdown 2026-07-10 21:23:33 -04:00
Magnus Hedemark bc04be7726 fix: derive risk severity from agent count instead of hardcoding medium 
RiskVector severity was hardcoded to 'medium' for all risks
regardless of how many agents flagged them. Now uses a simple
heuristic: 3+ agents = high, 2 agents = medium, 1 agent = low.
Also deduplicates similar risks and tracks which agents flagged
each one.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 01:51:50 -04:00
Magnus Hedemark 526e1ac3f0 fix: map AGENT_COUNCIL_API_KEY to provider-specific env var 
PydanticAI reads API keys from provider-specific env vars
(OPENAI_API_KEY, DEEPSEEK_API_KEY, etc.) at Agent creation time.
agent-council was reading AGENT_COUNCIL_API_KEY into a config dict
but never setting the env var PydanticAI actually looks for.

If a user had OPENAI_API_KEY set for something else, agent-council
silently used the wrong key for debate agents.

Fix: load_config() now maps AGENT_COUNCIL_API_KEY to the correct
env var based on the model prefix (openai:, deepseek:, anthropic:,
google:, etc.) and also maps AGENT_COUNCIL_BASE_URL to OPENAI_BASE_URL.
Only sets if not already set, so explicit env vars take precedence.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 01:45:07 -04:00
Magnus Hedemark 08d1011907 fix: update default model to openai:gpt-5.6-luna and fix model string format 
- Updates default model from gpt-4o-mini to gpt-5.6-luna across all
  documentation, config, and CLI help text
- Fixes model string format from provider/model to provider:model
  (PydanticAI convention) in README, SKILL.md, references, and examples
- Fixes anthropic, deepseek, and google example model strings to use
  colon format consistently

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 01:12:03 -04:00
Magnus Hedemark bddf955686 fix: add retries=3 to all debate agent phases 
Adds retry tolerance to structured output parsing across all
debate phases (premortem, position, cross-examine, synthesis,
compose, and verification guardrails). Prevents 'Exceeded
maximum output retries' failures when models produce
near-valid structured output on first attempt.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 01:04:17 -04:00
Magnus Hedemark 9a7f119ab0 fix: use enum notation for --agents and --mode in Command Reference 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 01:00:24 -04:00
Magnus Hedemark 6cb987073e fix: SkillOpt Epochs 1-3 — agent-council prominence, decision intelligence, and pattern expansion 
Epoch 1 — Prominence:
- Add --profiles flag to Command Reference
- Add Profile Selection section documenting 39 real profiles,
  auto-selection, and explicit --profiles selection
- Add Claims Verification section to Output documentation
- Add Pitfalls table for common issues

Epoch 2 — Decision Intelligence:
- Add Choosing a Profile Source decision table
- Add Reading the Convergence Diagnostic interpretation guide
  with 8 pattern→meaning→action mappings

Epoch 3 — Pattern Expansion:
- Add JSON output schema reference to Output section
- Add ai-frameworks and hermes-profiles to Related Skills

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 00:58:24 -04:00
Magnus Hedemark 9f3a68bd66 feat: replace fake personas with real profiles from hermes-profiles 
Instead of the compose phase fabricating personas with fake backgrounds,
the council now draws from 39 real professional profiles via a git
submodule (https://github.com/magnus919/hermes-profiles).

Key changes:
- New select phase reads SOUL.md + profile.yaml from profiles submodule
- Auto-updates submodule before selection via git submodule update --remote
- --profiles flag for explicit selection (comma-separated names)
- Auto-selection by keyword overlap with profile descriptions when omitted
- Each agent's identity is their real SOUL.md — actual methodology,
  values, and operating principles, not invented backgrounds
- Falls back to composed personas if profile library is unavailable
- Real profiles produce genuine methodological disagreement (debugger
  said 'unanswerable without a verified process' to naming question)

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-10 00:17:56 -04:00
Magnus Hedemark 5293ebd4d6 fix: add factuality guardrails and post-synthesis verification eval 
Two-layer defense against fabricated factual claims:

1. Prompt-level guardrail (FACTUAL_CLAIM_GUARDRAIL) injected into every
   debate agent's system prompt — tells agents not to assert specific
   verifiable facts they cannot know (domain availability, package
   namespace status, pricing, etc.)

2. Post-synthesis verification eval — an LLM-based QA pass that scans
   the narrative synthesis for unsubstantiated factual claims and
   appends a 'Claims Not Verified' section to the output with specific
   quotes, claim types, and explanations.

Both layers are self-contained — no search or external tools required.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 23:51:49 -04:00
Magnus Hedemark 6d4678cce5 fix: stream progress output during debate instead of silent stall 
- Each phase now prints completion status to stdout immediately
- Phase output files written to /tmp/agent-council/<ts>/ as they complete
- Users see real-time progress: persona count, phase timing,
  cross-examination metrics per round, convergence verdict
- Synthesis still printed at end as primary deliverable
- Full intermediate output available in /tmp/agent-council/ for inspection

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 23:21:16 -04:00
Magnus Hedemark e4ae4e4fd9 fix: add .env file support and update docs 
- Adds stdlib-only .env loader (no python-dotenv dependency)
- Env vars always take precedence over .env values
- Updates SKILL.md with .env usage example
- Updates configuration reference

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 23:05:12 -04:00
Magnus Hedemark 3a4aad40bf feat: add agent-council — multi-agent structured debate system 
Spawns a panel of expert agents to debate any question with
convergence-aware iteration and typed synthesis output.

- PydanticAI + PydanticGraph-based Python package
- 5-phase debate protocol: compose → premortem → position →
  cross-examine (iterative, eval-driven) → synthesis
- Convergence detection: confidence dispersion, argument
  novelty, concession rate — stops when diminishing returns
- Typed output schemas (Pydantic models) for every phase
- CLI tool with markdown and JSON output modes
- Custom persona file support
- Bootstrap detection: sys.executable -m pip install fallback
- agentskills.io compatible SKILL.md with triggers

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 22:55:30 -04:00
Magnus Hedemark 0c772a16b9 Add README.md requirement to AGENTS.md governance 
- Added 'Human-readable README' row to the Format Compliance table
- New README Format section defining 6 required sections + style guidance
- Added README validation to the Validate Your Output checklist
- References data-scientist/README.md as the canonical format example
 2026-07-09 22:36:49 -04:00
Magnus Hedemark 738ec715e7 Add human-focused README.md to every skill and bundle directory 
Each README is written for a human audience, explaining:
- What the skill does (not what format it follows)
- What benefit the user gets from installing it
- Quick setup and usage patterns
- When to load/trigger the skill
- What scripts, references, and templates it ships

data-scientist already had a README — left unchanged.

48 READMEs added across all skill and bundle directories.
 2026-07-09 22:30:12 -04:00
Magnus Hedemark 6c193ae2d8 enrich: 4 skills v1.1.0 — source validation, deepened references, worked examples 
DSPy v1.1.0: validation audit, worked RAG compilation example, expand ref table
Haystack v1.1.0: validation audit, file converters/YAML/component types, +2 refs
CrewAI v1.1.0: validation audit, unified Memory system, Flows docs, +3 refs
AutoGen v1.1.0: validation audit, v0.4 migration guide, AgentTool, streaming, +2 refs

All API surfaces validated against official docs.
 2026-07-09 15:15:17 -04:00
Magnus Hedemark 5428b18ddd feat: add autogen — expert skill for conversational multi-agent AI 
Greenfield SkillOpt: 3 epochs for Microsoft AutoGen skill.
Conversational agent model, GroupChat patterns, code execution,
nested chats, cancellation tokens, MCP integration.

All API surfaces validated against microsoft.github.io/autogen docs.
 2026-07-09 14:57:27 -04:00
Magnus Hedemark 48a67bb0a1 feat: add crewai — expert skill for role-based multi-agent teams 
Greenfield SkillOpt: 3 epochs for CrewAI skill.
Role/Goal/Backstory agent model, sequential/hierarchical processes.

11 files: SKILL.md, 6 references, 3 templates, 1 script.
 2026-07-09 14:55:42 -04:00
Magnus Hedemark fe5b275d00 feat: add haystack — expert skill for production search pipelines 
Greenfield SkillOpt: 3 epochs for deepset Haystack skill.
Pipeline DAG model, document stores, retrievers, evaluation, deployment.

Epoch 1 — Prominence: Hard-gate on Pipeline DAG vs LCEL pipe model
Epoch 2 — Decision Guidance: Where to Start, Framework Routing Guide
Epoch 3 — Pattern Expansion: Hybrid RAG pattern, evaluation pipeline, deployment

11 files: SKILL.md, 6 references, 3 templates, 1 script.
 2026-07-09 14:53:43 -04:00
Magnus Hedemark 95046675bc feat: add dspy — expert skill for compiled prompt programs 
Greenfield SkillOpt: 3 epochs for a Stanford DSPy compiler skill.
DSPy is a fundamentally different paradigm from chain/RAG frameworks.

Epoch 1 — Prominence:
- Hard-gate blockquote: 'DSPy is NOT a chain framework'
- Core Paradigm section with runnable code example early

Epoch 2 — Decision Guidance:
- Framework Routing Guide (DSPy vs LlamaIndex vs LangChain vs LangGraph)
- Where to Start table mapping entry points
- Troubleshooting table with reference links

Epoch 3 — Pattern Expansion:
- Optimizer selection cheat sheet from official docs
- Caching, compilation cost management, save/load
- FAQ covering paradigm confusion, errors, deployment

12 files: SKILL.md, 7 references, 3 templates, 1 script.
v1.0.0 -> v1.0.3 across 3 epochs.

All API surfaces validated against dspy.ai official docs —
optimizer selection guide, caching, core modules, FAQ.
Signed-off-by: Jasper <jasper@montcastle.bitches>
 2026-07-09 14:51:32 -04:00
Magnus Hedemark 7f2842b358 feat: langchain v1.1.0 — research-validated deepening 
Major deepening of the langchain expert skill based on source audit against
official LangChain docs (docs.langchain.com, reference.langchain.com).

Changes:
- Added references/validation-audit.md documenting all research findings
- Deepened references/agent-patterns.md from 74 to 200+ lines:
  create_react_agent full parameter table, @tool decorator with
  args_schema/parse_docstring, streaming events, multi-agent supervisor
- Deepened references/lcel-reference.md from 79 to 180+ lines:
  RunnablePassthrough.assign(), RunnableParallel dict shorthand,
  RunnableLambda, RunnableConfig, .with_fallbacks(), .configurable_fields()
- Deepened references/rag-strategies.md with advanced retrieval patterns
- Deepened references/production-deployment.md with LangSmith Datasets/
  Evaluation Runs/Prompt Hub
- Added new references/callbacks.md (BaseCallbackHandler, event table,
  agent auditing patterns, async callbacks)
- Deepened references/faq-and-troubleshooting.md with Pydantic v1/v2,
  streaming+tools, checkpoint serialization guidance

All API surface claims verified against official documentation.
v1.0.3 -> v1.1.0
 2026-07-09 14:33:42 -04:00
Magnus Hedemark fe3be63800 feat: add langchain — expert LangChain framework skill 
Greenfield SkillOpt: 3 epochs optimizing discoverability, decision
guidance, and troubleshooting for a brand-new LangChain skill.

Epoch 1 — Prominence:
- Added critical AgentExecutor deprecation callout at top
- Framework Routing Guide for cross-portfolio decisions

Epoch 2 — Decision Guidance:
- Where to Start table with AgentExecutor migration row
- Pipeline Mode table (Quick/RAG/Agent/Production)

Epoch 3 — Pattern Expansion:
- Troubleshooting table with reference file links
- FAQ section covering installation, migration, performance

13 files: SKILL.md, 7 references, 4 templates, 1 script.
v1.0.0 -> v1.0.3 across 3 SkillOpt epochs.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 14:13:34 -04:00
Magnus Hedemark c21ebb5afe feat: add llamaindex — expert LlamaIndex framework skill 
Greenfield SkillOpt: 3 epochs optimizing discoverability, decision guidance,
and pattern expansion for a brand-new LlamaIndex skill.

Epoch 1 — Prominence & Navigation:
- Moved Key Principles to top (before Quick Reference)
- Added Phase column to Quick Reference in workflow order
- Added 'When to use' column to Template Files
- Replaced flat When NOT to Use with Framework Routing Guide
  (LlamaIndex vs LangGraph vs PydanticAI vs Haystack vs DSPy)

Epoch 2 — Decision Guidance:
- Added Where to Start table (maps existing work → pipeline entry point)
- Added Pipeline Mode table (Quick/Full/Evaluate/Graph modes)

Epoch 3 — Pattern Expansion:
- Replaced flat gotchas with structured Troubleshooting Recovery Guide
  (3 symptom categories with immediate + permanent fixes)
- Added references/example-rag-pipeline.md (end-to-end worked example)
- Added references/evaluation-workflow.md (ParamTuner + evaluators)

17 files: SKILL.md, 11 references, 1 script, 4 templates.
v1.0.0 → v1.3.0 across 3 SkillOpt epochs.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 13:46:35 -04:00
Magnus Hedemark 3e86349539 feat: add llamaindex — expert skill for LlamaIndex framework 
Comprehensive skill covering:
- Core architecture (7 primitives, Settings, data flow)
- RAG strategies (basic through advanced with hybrid retrieval, reranking)
- Multi-agent orchestration (AgentWorkflow, handoff bug fix)
- Event-driven workflows (durable execution, checkpoint/resume)
- Production deployment (llama-deploy, debugging, observability)
- PropertyGraphIndex (knowledge graphs, hybrid retrieval)
- Evaluation and span-attached observability
- Integration ecosystem (vector stores, LlamaHub, LlamaParse)

9 reference files, 4 templates, 1 verification script.
MIT licensed. 100% AI agent portable (no platform-specific content).

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 13:32:53 -04:00
Magnus Hedemark 783bff4a03 chore: remove stale validation artifact 
Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 05:55:50 -04:00
Magnus Hedemark f3ebdedc87 fix: SkillOpt Epoch 3 — pydanticai framework boundaries 
Edits accepted and merged:
- Framework comparison table: PydanticAI vs LangGraph vs using both together
- Boundary conditions: when to choose each framework
- Hybrid pattern reference: PydanticAI agent as LangGraph StateGraph node
- Updated 'When to Load Which Reference' table with boundaries entry
- Updated Directory Structure listing

New reference file: references/hybrid-pydanticai-langgraph.md (7KB)
Version bumped from 1.0.3 to 1.0.4.

All 3 validation tasks passed with no regressions:
- Val-1: Multi-agent delegation (pass)
- Val-2: Hybrid PydanticAI+LangGraph pattern (pass, all 6 criteria)
- Val-3: Streaming agent (pass)

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 05:55:41 -04:00
Magnus Hedemark 9ff931ddf5 fix: SkillOpt Epoch 2 — pydanticai decision intelligence 
Edits accepted and merged:
- Run method decision table (when to use run/run_sync/run_stream/run_stream_events/iter)
- Graph API comparison table (BaseNode vs GraphBuilder trade-offs)
- Error handling quick-pick with exception table and recovery patterns

Version bumped from 1.0.2 to 1.0.3.

All 3 validation tasks passed with no regressions (6/6, 10/10, 8/8 rubric items).

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 05:48:35 -04:00
Magnus Hedemark 3cfdcf8677 fix: SkillOpt Epoch 1 — pydanticai optimization 
Edits accepted and merged:
- Added defer_model_check gotcha to SKILL.md Gotchas section
- Added defer_model_check usage note to core-agents.md constructor params
- Added pytest-asyncio dependency note to testing-evals.md
- stream_text/structured output gotcha added from validation findings
- graph.run() returns output not state gotcha added from validation findings

Version bumped from 1.0.0 to 1.0.1.

All 3 training rollouts passed (6/6, 5/5, 6/6 rubric items).
All 3 validation tasks passed (6/6, 10/10, 8/8 rubric items).
No regressions detected.

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-09 05:42:24 -04:00
Magnus Hedemark e4a11880d1 feat: add pydanticai skill — PydanticAI & PydanticGraph expert reference 
Comprehensive agent skill covering:
- Agent creation, function tools, dependency injection, instructions
- 20+ capability system with on-demand (deferred) loading
- Lifecycle hooks system (before/after/wrap for all phases)
- 16 model providers, FallbackModel, ConcurrencyLimitedModel
- Structured output, streaming, output functions
- Multi-agent delegation and programmatic hand-off
- PydanticGraph: both BaseNode (class-based) and GraphBuilder (function-based)
  with parallel execution, joins/reducers, decisions, Mermaid rendering
- Testing with TestModel/FunctionModel and eval framework
- MCP integration, durable execution, UI adapters
- 8 comprehensive reference files + API quick reference

Signed-off-by: Magnus Hedemark <magnus919@pm.me>
 2026-07-08 23:50:36 -04:00
Magnus Hedemark e3bd060de0 fix: SkillOpt Epoch 2 — langgraph decision intelligence 
Epoch 2 edits (3/3 accepted, 3/3 validation passed):
- ADD 'Start from template' column to Pattern Selection Guide
- ADD 'Choosing Your Starting Point' table (scaffold vs template vs reference)
- ADD concrete template guidance to Quick Start next-steps

Version: 1.0.1 → 1.0.2
 2026-07-08 15:17:25 -04:00
Magnus Hedemark 2731a3a48f fix: SkillOpt Epoch 1 — langgraph prominence and discoverability 
Epoch 1 edits (4/4 accepted, 3/3 validation tasks passed):
- ADD Quick Start section with pip install + hello-world + next-steps
- MOVE 6 design principles from architecture.md into SKILL.md as hard-gate blockquote
- ADD 3 template rows to Reference Files table
- ADD install dependencies blockquote at top

Version: 1.0.0 → 1.0.1
 2026-07-08 15:08:03 -04:00
Magnus Hedemark 4a73657522 feat: add langgraph expert skill — multi-agent patterns, scaffolds, evals, and production guidance 
Comprehensive LangGraph skill covering:
- Core architecture: Graph API, Functional API, state management, agent loops
- Three multi-agent patterns: supervisor (~94% accuracy), swarm (~40% fewer LLM calls),
  hierarchical teams (subgraphs with nested state)
- Persistence: checkpointers vs stores, per-invocation/per-thread/stateless modes
- Production: Agent Server deployment, LangSmith observability, 8 failure modes
- Evals: routing accuracy, resolution coverage, LLM-as-judge methodology
- Troubleshooting: symptom→cause→fix tables per pattern
- 3 Python scripts: supervisor scaffold, swarm scaffold, eval generator
- 3 runnable templates: supervisor, swarm, subgraph composition

Ships 8 reference files, 3 scripts, and 3 templates.
 2026-07-08 14:54:53 -04:00
Magnus Hedemark 1275b7dabc fix: consolidate research subagent content into product-discovery references 
Adds Socratic questioning (6 types), tacit knowledge extraction
(master-apprentice, think-aloud, critical incident, artifact walkthrough),
interview protocol design table, The Mom Test framework, expanded
non-defensive follow-up techniques, and practitioner reference table.
Consolidates subagent-created skills into the main skill's reference files.
 2026-07-06 19:42:22 -04:00
Magnus Hedemark 3f03c6e327 feat: add product-discovery skill — Phase 0 methodology for requirements discovery upstream of SDD 
Covers stakeholder mapping, question patterns, gap detection, conflict
resolution, transcript-to-spec distillation, AI-conducted discovery,
power dynamics, and time-constrained discovery. Ships 8 references and
5 templates.
 2026-07-06 13:29:44 -04:00
Magnus Hedemark 980f87050d feat: spec-driven-development v1.2.0 — SDD methodology for AI software factories 2026-07-06 00:09:07 -04:00
Magnus Hedemark 920c496cc5 fix: bump crowdsec version to 0.0.2 2026-07-05 22:34:04 -04:00
Magnus Hedemark d438952dfc fix: SkillOpt Epoch 2 — crowdsec skill optimization 
- Expand Nginx bouncer section with full config example and directives
- Add nginx-bouncer reference file with Cloudflare/CDN guidance
- Add flush tuning guidance (recommended values, VACUUM, low-power tuning)
- Add missing cscli subcommands to summary table (config, explain, simulation, allowlists)
- Clean up redundant gotchas entries
 2026-07-05 22:32:18 -04:00
Magnus Hedemark 8648c59452 fix: SkillOpt Epoch 1 — crowdsec skill optimization 
- Add version field (0.0.1)
- Tighten description from 317 to ~240 chars
- Fix Quick Reference: decisions remove -> decisions delete
- Extract cscli command reference to dedicated reference file
- Add poll_without_inotify note to Acquisition section
 2026-07-05 22:23:18 -04:00