magnus919_agent-skills/docs/lifecycle-capability-matrix.md

# Lifecycle Capability Matrix

Deterministically generated by `scripts/gen-lifecycle-matrix.rb` (schema version 1). 
Every matrix entry derives from a declared bundle manifest field or a documented 
derivation; see [bundle-manifest-design.md](bundle-manifest-design.md) for the schema, 
the migration path, and the field definitions. Regenerate with 
`ruby scripts/gen-lifecycle-matrix.rb --write`.

## Summary

| Bundle | Purpose | Audience | Lifecycle stages | Included skills | Prerequisites | Outputs | Handoffs | Conflicts | Eval suite |
|---|---|---|---|---|---|---|---|---|---|
| agent-production-operations | Operate an evaluated agent with tools and authority in production through a runtime control plane covering versioning, staged rollout, fallback, cost and latency budgets, tool health, human escalation, disablement, and trace-to-eval feedback. | Agent operators and platform teams running evaluated agents with tools and delegated authority in production; agents that need the runtime control decisions between evaluation, release, SRE, security, and platform specialists. | Contract definition → Versioning and rollout planning → Staged rollout → Runtime monitoring → Trace-to-eval feedback → Learning and disablement | agent-evals-and-observability, release-engineering, site-reliability-engineering, secure-software-engineering, platform-engineering, production-readiness, incident-learning, privacy-engineering | Agent evaluation results and observability instrumentation(agent-evals-and-observability); Production-readiness evidence (go/no-go/defer/exception outcomes)(production-readiness); Incident-learning records (verified closure, follow-up work map)(incident-learning); Release pipeline and artifact promotion baseline(release-engineering) | agent-production-contract, runtime-control-plan, staged-rollout-plan, tool-authority-health-record, trace-to-eval-feedback-loop, disablement-record | staged-rollout-plan → release-engineering; trace-to-eval-feedback-loop → agent-evals-and-observability; tool-authority-health-record → incident-learning; disablement-record → human operator | production-readiness (with production-excellence); production-readiness (with product-lifecycle); release-engineering (with production-excellence); release-engineering (with product-lifecycle); site-reliability-engineering (with production-excellence); site-reliability-engineering (with product-lifecycle); secure-software-engineering (with production-excellence); secure-software-engineering (with product-lifecycle); platform-engineering (with production-excellence); platform-engineering (with product-lifecycle); incident-learning (with production-excellence); privacy-engineering (with product-lifecycle) | evals/evals.json |
| neckbeard | A disciplined SDLC delivery operating model for non-trivial software change — fix, build, refactor, review, verify, or release — that routes the work through framing, discovery, design, implementation, review, verification, delivery, and learning, choosing the smallest *safe* intervention, proving it at the real delivery boundary, and leaving an inspectable evidence ledger. | Software engineers and agents delivering non-trivial software changes (bug fixes, features, refactors, reviews, releases) that need a bounded, stage-aware path to a defensible 'done' with recorded evidence; teams running issue-to-PR change-request journeys with gates and an authorized release. | Intake and framing → Current-state discovery → Design → Specification → Test planning → Implementation → Review and verification → Readiness and release | product-discovery, software-architecture-analysis, systematic-debugging, api-design-and-evolution, product-design-and-ux, spec-driven-development, qa-methodology, secure-software-engineering, web-accessibility, verification-methodology, technical-documentation, release-engineering, site-reliability-engineering | A non-trivial change request, issue, or task with a verifiable contract; Current-state evidence (repository, tests, config, recent history)(software-architecture-analysis); Change contract and evidence ledger entries(verification-methodology) | change-contract, evidence-ledger, delivery-packet, implementation-plan, verification-report, release-record | verification-report → verification-methodology; release-record → release-engineering; delivery-packet → human reviewer; evidence-ledger → next change (via evidence ledger) | product-discovery (with product-lifecycle); product-design-and-ux (with product-lifecycle); spec-driven-development (with product-lifecycle); qa-methodology (with product-lifecycle); qa-methodology (with production-excellence); secure-software-engineering (with product-lifecycle); secure-software-engineering (with production-excellence); secure-software-engineering (with agent-production-operations); verification-methodology (with product-lifecycle); verification-methodology (with production-excellence); release-engineering (with product-lifecycle); release-engineering (with production-excellence); release-engineering (with agent-production-operations); site-reliability-engineering (with product-lifecycle); site-reliability-engineering (with production-excellence); site-reliability-engineering (with agent-production-operations) | evals/evals.json |
| product-lifecycle | Route a product through its full lifecycle — discovery, strategy and portfolio choice, roadmap, UX and requirements, experimentation, delivery handoff, adoption, success, and lifecycle review — by composing existing specialist product skills with phase-entry evidence, handoff artifacts, and escalation rules. | Product managers, product operations, and portfolio teams navigating a product across multiple lifecycle phases; agents that need to know which specialist skill to load at each phase and what evidence to hand off to the next phase. | Discovery → Strategy and portfolio choice → Roadmap → UX and requirements → Experimentation → Delivery handoff → Adoption → Success → Lifecycle review | product-discovery, product-strategy, strategy-frameworks, product-roadmapping-and-portfolio, product-design-and-ux, product-experimentation, implementation-planning, production-readiness, release-engineering, product-adoption, product-analytics-and-measurement, conditional-customer-success, product-lifecycle-learning, product-methodology, product-operations-and-governance, financial-modeling, go-to-market, data-scientist, spec-driven-development, privacy-engineering, secure-software-engineering, verification-methodology, qa-methodology, site-reliability-engineering, platform-engineering, neckbeard | Product idea, market signal, or stakeholder request; Problem statement and discovery log(product-discovery); Strategic assessment and portfolio decision(product-strategy); Roadmap entry and bet record(product-roadmapping-and-portfolio); UX contracts and acceptance criteria(product-design-and-ux); Experiment readout and assumptions register(product-experimentation) | lifecycle-evidence-ledger, problem-statement, strategic-assessment, roadmap-entry, interface-contracts, experiment-brief, implementation-plan, readiness-verdict, release-plan, adoption-plan, outcome-measurement, outcome-review, lifecycle-decision | readiness-verdict → production-excellence bundle; implementation-plan → delivery team; lifecycle-evidence-ledger → next phase (via evidence ledger); lifecycle-decision → product-lifecycle-learning | production-readiness (with production-excellence); production-readiness (with agent-production-operations); release-engineering (with production-excellence); release-engineering (with agent-production-operations); site-reliability-engineering (with production-excellence); site-reliability-engineering (with agent-production-operations); secure-software-engineering (with production-excellence); secure-software-engineering (with agent-production-operations); platform-engineering (with production-excellence); platform-engineering (with agent-production-operations); privacy-engineering (with agent-production-operations); qa-methodology (with production-excellence); verification-methodology (with production-excellence) | evals/evals.json |
| production-excellence | Route cross-domain production evidence (readiness, migration, recovery, capacity/cost, incident-learning) into a launch or operational decision — go, no-go, defer, exception, or escalation — with an accountable owner and a post-launch learning path. | Service owners, SRE, release, platform, security, data, and QA teams coordinating a production change; agents assembling cross-domain production evidence into a defensible launch decision without re-deriving specialist runbooks. | Cross-domain evidence assembly → Gate evaluation → Operational handoff → Post-launch learning | production-readiness, migration-engineering, resilience-and-recovery, capacity-and-cost-engineering, incident-learning, site-reliability-engineering, release-engineering, platform-engineering, secure-software-engineering, data-engineering, qa-methodology, verification-methodology | Readiness review evidence packet (risk-scaled, 11 categories)(production-readiness); Migration plan, compatibility windows, and reconciliation evidence(migration-engineering); Resilience and recovery evidence (RTO/RPO, restore tests, game days)(resilience-and-recovery); Capacity and cost model evidence (demand, scaling, unit cost)(capacity-and-cost-engineering); Incident history, causal hypotheses, and verified-closure records(incident-learning); SLO and error-budget status(site-reliability-engineering); Release plan and rollback strategy(release-engineering) | gate-decision, operational-handoff-record, accountable-owner-record, evidence-gap-register | operational-handoff-record → incident-learning; operational-handoff-record → product-lifecycle-learning; gate-decision → agent-production-operations; accountable-owner-record → service owner | production-readiness (with product-lifecycle); production-readiness (with agent-production-operations); release-engineering (with product-lifecycle); release-engineering (with agent-production-operations); site-reliability-engineering (with product-lifecycle); site-reliability-engineering (with agent-production-operations); secure-software-engineering (with product-lifecycle); secure-software-engineering (with agent-production-operations); platform-engineering (with product-lifecycle); platform-engineering (with agent-production-operations); incident-learning (with agent-production-operations); qa-methodology (with product-lifecycle); verification-methodology (with product-lifecycle) | evals/evals.json |
| research-and-vault | Chain web research, atomic extraction, and durable knowledge capture into a repeatable research-to-note workflow: capture source URLs and dates before interpretation, extract independent atomic claims before synthesis, and write the durable note only after source coverage and gaps are explicit. | Agents and knowledge workers that repeatedly need the full research-to-note sequence (web research, atomic extraction, durable capture) and want a stable order of operations with a clear handoff artifact instead of ad-hoc steps; not for single lookups or single notes, which route to the domain skill directly. | Web research → Source capture → Atomic extraction → Synthesis and durable capture | research-methodology | A research question or topic and access to web-research tooling(research-methodology) | source-capture-record, atomic-claim-extract, durable-note | source-capture-record → research-methodology; atomic-claim-extract → research-methodology; durable-note → user | none declared | evals/evals.json |
| tailscale | Operate a self-hosted Tailscale/Headscale ecosystem end to end: deploy and manage a Headscale control server, configure Tailscale clients, define ACL policy, manage node lifecycle, advertise subnet routes and exit nodes, operate DERP relays, and back up, restore, or migrate the control plane. | Platform and homelab operators running self-hosted Tailscale/Headscale infrastructure; agents that need to know which sub-skill to load for deploy, policy, client, node, routing, DERP, or backup tasks in a WireGuard mesh. | Deploy control server → Policy → Client connectivity → Node lifecycle → Routing → DERP relays → Backup and restore | headscale-deploy, tailnet-policy, tailscale-client, headscale-node-lifecycle, headscale-routing, headscale-derp, headscale-backup | A Linux host and install target for the Headscale control server(headscale-deploy); HEADSCALE_URL and HEADSCALE_API_KEY credentials(headscale-deploy); A running Headscale instance with the headscale CLI available(headscale-deploy) | headscale-server, tailnet-policy, registered-nodes, routed-networks, derp-map, backup-archive, restored-instance | headscale-server → tailnet-policy; headscale-server → headscale-node-lifecycle; headscale-server → headscale-backup; backup-archive → user | none declared | evals/evals.json |
| workflow-architect | Discover a user's actual workflow through active interrogation or passive observation, then generate a tailored Agent Skills bundle that encodes it as loadable skills with trigger conditions, a decision map, a manifest, and an optional kanban board — turning a session that feels aimless into structure. | Users and agents that want to formalize how they work — understand their own process, encode it as trigger-driven skills they can reuse, or share it with collaborators. Not for single factual questions or tasks already owned by a more specific skill. | Active interview → Passive observation → Synthesis and generation | interviewer, observer, bundle-builder | Session context with enough turns for observation (or user answers for interview)(observer); Convergence on workflow phases, branching signals, and tool preferences(interviewer) | skills-bundle, decision-map, manifest, kanban-board | skills-bundle → user; decision-map → user; manifest → user; kanban-board → user | none declared | evals/evals.json |

## Details

### agent-production-operations

- Derivation: `manifest` — source: `bundles/agent-production-operations/manifest.yaml`
- **Purpose:** Operate an evaluated agent with tools and authority in production through a runtime control plane covering versioning, staged rollout, fallback, cost and latency budgets, tool health, human escalation, disablement, and trace-to-eval feedback.
- **Audience:** Agent operators and platform teams running evaluated agents with tools and delegated authority in production; agents that need the runtime control decisions between evaluation, release, SRE, security, and platform specialists.
- **Stages (ordered):**
  1. Contract definition — agent-evals-and-observability (`../../agent-evals-and-observability/SKILL.md`), production-readiness (`../../production-readiness/SKILL.md`), incident-learning (`../../incident-learning/SKILL.md`)
  2. Versioning and rollout planning — release-engineering (`../../release-engineering/SKILL.md`), platform-engineering (`../../platform-engineering/SKILL.md`)
  3. Staged rollout — platform-engineering (`../../platform-engineering/SKILL.md`), site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
  4. Runtime monitoring — site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`), secure-software-engineering (`../../secure-software-engineering/SKILL.md`)
  5. Trace-to-eval feedback — agent-evals-and-observability (`../../agent-evals-and-observability/SKILL.md`)
  6. Learning and disablement — incident-learning (`../../incident-learning/SKILL.md`), production-readiness (`../../production-readiness/SKILL.md`), privacy-engineering (`../../privacy-engineering/SKILL.md`)
- **Included skills:**
  - agent-evals-and-observability (`../../agent-evals-and-observability/SKILL.md`)
  - release-engineering (`../../release-engineering/SKILL.md`)
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`)
  - platform-engineering (`../../platform-engineering/SKILL.md`)
  - production-readiness (`../../production-readiness/SKILL.md`)
  - incident-learning (`../../incident-learning/SKILL.md`)
  - privacy-engineering (`../../privacy-engineering/SKILL.md`)
- **Prerequisites:**
  - Agent evaluation results and observability instrumentation — via agent-evals-and-observability (`../../agent-evals-and-observability/SKILL.md`)
  - Production-readiness evidence (go/no-go/defer/exception outcomes) — via production-readiness (`../../production-readiness/SKILL.md`)
  - Incident-learning records (verified closure, follow-up work map) — via incident-learning (`../../incident-learning/SKILL.md`)
  - Release pipeline and artifact promotion baseline — via release-engineering (`../../release-engineering/SKILL.md`)
- **Outputs:** agent-production-contract, runtime-control-plan, staged-rollout-plan, tool-authority-health-record, trace-to-eval-feedback-loop, disablement-record
- **Handoffs:**
  - staged-rollout-plan → release-engineering — Rollout stages and promotion gates integrate with the release pipeline; rollback triggers tie agent health to release mechanics.
  - trace-to-eval-feedback-loop → agent-evals-and-observability — Production traces and incidents feed new evaluation cases and release-gate updates.
  - tool-authority-health-record → incident-learning — Tool-failure and authority-breach records feed post-incident analysis and verified-closure tracking.
  - disablement-record → human operator — Disablement revokes credentials, removes the agent from routing/load-balancing, and records the reason, timestamp, and authorizing evidence.
- **Conflicts:**
  - production-readiness (`../../production-readiness/SKILL.md`) with production-excellence — Both bundles include production-readiness. This bundle consumes readiness outcomes as authority-gating inputs; production-excellence produces the go/no-go/defer/exception decisions.
  - production-readiness (`../../production-readiness/SKILL.md`) with product-lifecycle — Both bundles include production-readiness. Route phase 6 delivery readiness reviews to product-lifecycle; runtime authority decisions for an agent in production go to this bundle.
  - release-engineering (`../../release-engineering/SKILL.md`) with production-excellence — Release engineering is shared. Route by context: agent staged-rollout planning (this bundle) versus launch-gate release plans (production-excellence).
  - release-engineering (`../../release-engineering/SKILL.md`) with product-lifecycle — Release engineering is shared. Route by context: agent staged-rollout planning (this bundle) versus lifecycle delivery handoff (product-lifecycle).
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with production-excellence — SRE is shared. This bundle uses it for runtime latency/cost budgets and incident handoff; production-excellence uses it for gate-entry reliability evidence. Incident command stays with SRE in both cases.
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with product-lifecycle — SRE is shared. This bundle uses it for agent runtime budgets; product-lifecycle routes to it for delivery reliability.
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with production-excellence — Security engineering is shared. Route by context: agent authority contracts and disablement security (this bundle) versus launch-gate security evidence (production-excellence).
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with product-lifecycle — Security engineering is shared. Route by context: agent trust-boundary and disablement security (this bundle) versus lifecycle-phase security needs (product-lifecycle).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with production-excellence — Platform engineering is shared. Route by context: agent sandboxing and tool-execution isolation (this bundle) versus launch-gate platform evidence (production-excellence).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with product-lifecycle — Platform engineering is shared. Route by context: agent sandboxing and tool-execution isolation (this bundle) versus lifecycle delivery handoff (product-lifecycle).
  - incident-learning (`../../incident-learning/SKILL.md`) with production-excellence — Both bundles consume incident-learning records. This bundle consumes verified-closure records as escalation and disablement inputs; production-excellence uses incident history as gate-entry evidence.
  - privacy-engineering (`../../privacy-engineering/SKILL.md`) with product-lifecycle — Privacy engineering is shared. This bundle routes to it for trace scrubbing and user-data-access grants; product-lifecycle routes to it for PII, consent, retention, and data flows in any phase.
- **Eval suite:** evals/evals.json

### neckbeard

- Derivation: `manifest` — source: `bundles/neckbeard/manifest.yaml`
- **Purpose:** A disciplined SDLC delivery operating model for non-trivial software change — fix, build, refactor, review, verify, or release — that routes the work through framing, discovery, design, implementation, review, verification, delivery, and learning, choosing the smallest *safe* intervention, proving it at the real delivery boundary, and leaving an inspectable evidence ledger.
- **Audience:** Software engineers and agents delivering non-trivial software changes (bug fixes, features, refactors, reviews, releases) that need a bounded, stage-aware path to a defensible 'done' with recorded evidence; teams running issue-to-PR change-request journeys with gates and an authorized release.
- **Stages (ordered):**
  1. Intake and framing — product-discovery (`../../product-discovery/SKILL.md`)
  2. Current-state discovery — software-architecture-analysis (`../../software-architecture-analysis/SKILL.md`), systematic-debugging (`../../systematic-debugging/SKILL.md`)
  3. Design — api-design-and-evolution (`../../api-design-and-evolution/SKILL.md`), product-design-and-ux (`../../product-design-and-ux/SKILL.md`)
  4. Specification — spec-driven-development (`../../spec-driven-development/SKILL.md`)
  5. Test planning — qa-methodology (`../../qa-methodology/SKILL.md`)
  6. Implementation — secure-software-engineering (`../../secure-software-engineering/SKILL.md`), web-accessibility (`../../web-accessibility/SKILL.md`)
  7. Review and verification — verification-methodology (`../../verification-methodology/SKILL.md`), technical-documentation (`../../technical-documentation/SKILL.md`)
  8. Readiness and release — release-engineering (`../../release-engineering/SKILL.md`), site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
- **Included skills:**
  - product-discovery (`../../product-discovery/SKILL.md`)
  - software-architecture-analysis (`../../software-architecture-analysis/SKILL.md`)
  - systematic-debugging (`../../systematic-debugging/SKILL.md`)
  - api-design-and-evolution (`../../api-design-and-evolution/SKILL.md`)
  - product-design-and-ux (`../../product-design-and-ux/SKILL.md`)
  - spec-driven-development (`../../spec-driven-development/SKILL.md`)
  - qa-methodology (`../../qa-methodology/SKILL.md`)
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`)
  - web-accessibility (`../../web-accessibility/SKILL.md`)
  - verification-methodology (`../../verification-methodology/SKILL.md`)
  - technical-documentation (`../../technical-documentation/SKILL.md`)
  - release-engineering (`../../release-engineering/SKILL.md`)
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
- **Prerequisites:**
  - A non-trivial change request, issue, or task with a verifiable contract
  - Current-state evidence (repository, tests, config, recent history) — via software-architecture-analysis (`../../software-architecture-analysis/SKILL.md`)
  - Change contract and evidence ledger entries — via verification-methodology (`../../verification-methodology/SKILL.md`)
- **Outputs:** change-contract, evidence-ledger, delivery-packet, implementation-plan, verification-report, release-record
- **Handoffs:**
  - verification-report → verification-methodology — The verification report records the declared verification target and the boundary actually exercised; verdicts bind to the exact head and never claim a boundary that was not checked.
  - release-record → release-engineering — Release planning, versioning, pipeline promotion, readiness, and authorized release routing for the delivered change.
  - delivery-packet → human reviewer — Change-request journeys hand the delivery packet — gates and evidence — to the required review before merge; a timed-out review is inconclusive, not a reason to launch repeated review rounds.
  - evidence-ledger → next change (via evidence ledger) — Each run emits a compact evidence record (intent, inspected artifacts, assumptions, rejected alternatives, checks run, observed outputs, unverified boundaries) that carries learning into the next change.
- **Conflicts:**
  - product-discovery (`../../product-discovery/SKILL.md`) with product-lifecycle — Product discovery is shared. Route by context: neckbeard loads it for intake and framing of a change request; product-lifecycle owns the full discovery phase of a product lifecycle.
  - product-design-and-ux (`../../product-design-and-ux/SKILL.md`) with product-lifecycle — Product design is shared. neckbeard loads it for user-facing behavior, interaction, and information architecture during design; product-lifecycle owns the UX and requirements phase of a product.
  - spec-driven-development (`../../spec-driven-development/SKILL.md`) with product-lifecycle — Specification is shared. neckbeard uses it for formal specification and phase gates on a change request; product-lifecycle uses it for requirements and acceptance criteria in its lifecycle phases.
  - qa-methodology (`../../qa-methodology/SKILL.md`) with product-lifecycle — QA methodology is shared. neckbeard loads it for test strategy and regression gates on a change; product-lifecycle routes to it for phase 6 quality gates.
  - qa-methodology (`../../qa-methodology/SKILL.md`) with production-excellence — QA methodology is shared. neckbeard loads it for change-request test planning; production-excellence uses it for launch verification evidence.
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with product-lifecycle — Security engineering is shared. Route by context: change-level security review and threat modeling (neckbeard) versus lifecycle-phase security needs (product-lifecycle).
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with production-excellence — Security engineering is shared. Route by context: change-level security review (neckbeard) versus launch-gate security evidence (production-excellence).
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with agent-production-operations — Security engineering is shared. Route by context: change-level security review (neckbeard) versus agent authority contracts and disablement security (agent-production-operations).
  - verification-methodology (`../../verification-methodology/SKILL.md`) with product-lifecycle — Verification methodology is shared. neckbeard requires it to exercise the declared verification boundary on a change; product-lifecycle uses it at every phase gate for evidence.
  - verification-methodology (`../../verification-methodology/SKILL.md`) with production-excellence — Verification methodology is shared. neckbeard uses it for change-request verification verdicts; production-excellence uses it for boundary labeling of launch evidence.
  - release-engineering (`../../release-engineering/SKILL.md`) with product-lifecycle — Release engineering is shared. Route by context: change-request release planning (neckbeard) versus lifecycle delivery handoff (product-lifecycle).
  - release-engineering (`../../release-engineering/SKILL.md`) with production-excellence — Release engineering is shared. Route by context: change-request release planning (neckbeard) versus launch-gate release evidence (production-excellence).
  - release-engineering (`../../release-engineering/SKILL.md`) with agent-production-operations — Release engineering is shared. Route by context: change-request release planning (neckbeard) versus agent staged-rollout planning (agent-production-operations).
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with product-lifecycle — SRE is shared. neckbeard routes to it for reliability and operational recovery of a delivered change; product-lifecycle routes to it for delivery reliability.
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with production-excellence — SRE is shared. Route by context: change-level reliability (neckbeard) versus gate-entry SLO/error-budget evidence (production-excellence).
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with agent-production-operations — SRE is shared. Route by context: change-level reliability (neckbeard) versus agent runtime latency/cost budgets and incident handoff (agent-production-operations).
- **Eval suite:** evals/evals.json

### product-lifecycle

- Derivation: `manifest` — source: `bundles/product-lifecycle/manifest.yaml`
- **Purpose:** Route a product through its full lifecycle — discovery, strategy and portfolio choice, roadmap, UX and requirements, experimentation, delivery handoff, adoption, success, and lifecycle review — by composing existing specialist product skills with phase-entry evidence, handoff artifacts, and escalation rules.
- **Audience:** Product managers, product operations, and portfolio teams navigating a product across multiple lifecycle phases; agents that need to know which specialist skill to load at each phase and what evidence to hand off to the next phase.
- **Stages (ordered):**
  1. Discovery — product-discovery (`../../product-discovery/SKILL.md`)
  2. Strategy and portfolio choice — product-strategy (`../../product-strategy/SKILL.md`), strategy-frameworks (`../../strategy-frameworks/SKILL.md`)
  3. Roadmap — product-roadmapping-and-portfolio (`../../product-roadmapping-and-portfolio/SKILL.md`)
  4. UX and requirements — product-design-and-ux (`../../product-design-and-ux/SKILL.md`)
  5. Experimentation — product-experimentation (`../../product-experimentation/SKILL.md`)
  6. Delivery handoff — implementation-planning (`../../implementation-planning/SKILL.md`), production-readiness (`../../production-readiness/SKILL.md`), release-engineering (`../../release-engineering/SKILL.md`)
  7. Adoption — product-adoption (`../../product-adoption/SKILL.md`)
  8. Success — product-analytics-and-measurement (`../../product-analytics-and-measurement/SKILL.md`), conditional-customer-success (`../../conditional-customer-success/SKILL.md`)
  9. Lifecycle review — product-lifecycle-learning (`../../product-lifecycle-learning/SKILL.md`)
- **Included skills:**
  - product-discovery (`../../product-discovery/SKILL.md`)
  - product-strategy (`../../product-strategy/SKILL.md`)
  - strategy-frameworks (`../../strategy-frameworks/SKILL.md`)
  - product-roadmapping-and-portfolio (`../../product-roadmapping-and-portfolio/SKILL.md`)
  - product-design-and-ux (`../../product-design-and-ux/SKILL.md`)
  - product-experimentation (`../../product-experimentation/SKILL.md`)
  - implementation-planning (`../../implementation-planning/SKILL.md`)
  - production-readiness (`../../production-readiness/SKILL.md`)
  - release-engineering (`../../release-engineering/SKILL.md`)
  - product-adoption (`../../product-adoption/SKILL.md`)
  - product-analytics-and-measurement (`../../product-analytics-and-measurement/SKILL.md`)
  - conditional-customer-success (`../../conditional-customer-success/SKILL.md`)
  - product-lifecycle-learning (`../../product-lifecycle-learning/SKILL.md`)
  - product-methodology (`../../product-methodology/SKILL.md`)
  - product-operations-and-governance (`../../product-operations-and-governance/SKILL.md`)
  - financial-modeling (`../../financial-modeling/SKILL.md`)
  - go-to-market (`../../go-to-market/SKILL.md`)
  - data-scientist (`../../data-scientist/SKILL.md`)
  - spec-driven-development (`../../spec-driven-development/SKILL.md`)
  - privacy-engineering (`../../privacy-engineering/SKILL.md`)
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`)
  - verification-methodology (`../../verification-methodology/SKILL.md`)
  - qa-methodology (`../../qa-methodology/SKILL.md`)
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
  - platform-engineering (`../../platform-engineering/SKILL.md`)
  - neckbeard (`../../bundles/neckbeard/SKILL.md`)
- **Prerequisites:**
  - Product idea, market signal, or stakeholder request
  - Problem statement and discovery log — via product-discovery (`../../product-discovery/SKILL.md`)
  - Strategic assessment and portfolio decision — via product-strategy (`../../product-strategy/SKILL.md`)
  - Roadmap entry and bet record — via product-roadmapping-and-portfolio (`../../product-roadmapping-and-portfolio/SKILL.md`)
  - UX contracts and acceptance criteria — via product-design-and-ux (`../../product-design-and-ux/SKILL.md`)
  - Experiment readout and assumptions register — via product-experimentation (`../../product-experimentation/SKILL.md`)
- **Outputs:** lifecycle-evidence-ledger, problem-statement, strategic-assessment, roadmap-entry, interface-contracts, experiment-brief, implementation-plan, readiness-verdict, release-plan, adoption-plan, outcome-measurement, outcome-review, lifecycle-decision
- **Handoffs:**
  - readiness-verdict → production-excellence bundle — Phase 6 delivery handoff routes the production-readiness verdict to the production-excellence launch gate when that bundle is available.
  - implementation-plan → delivery team — Phase 6 hands the implementation plan and release plan to the delivery team (and neckbeard for SDLC change-request work).
  - lifecycle-evidence-ledger → next phase (via evidence ledger) — Every phase reads and writes the shared lifecycle evidence ledger; the next phase never re-derives prior-phase evidence.
  - lifecycle-decision → product-lifecycle-learning — Phase 9 records the continue/improve/harvest/pivot/pause/retire decision and retained learning in the learning loop.
- **Conflicts:**
  - production-readiness (`../../production-readiness/SKILL.md`) with production-excellence — Both bundles include production-readiness. Use production-excellence for launch-gate decisions; use product-lifecycle for phase 6 delivery handoff routing. Load the bundle matching the active decision context, never both at once.
  - production-readiness (`../../production-readiness/SKILL.md`) with agent-production-operations — agent-production-operations consumes readiness outcomes as authority- gating inputs. Route phase 6 readiness reviews here; runtime authority decisions go to agent-production-operations.
  - release-engineering (`../../release-engineering/SKILL.md`) with production-excellence — Release engineering is shared. Route by context: lifecycle delivery handoff (this bundle) versus gate-entry release evidence (production-excellence).
  - release-engineering (`../../release-engineering/SKILL.md`) with agent-production-operations — Release engineering is shared. Route by context: lifecycle delivery handoff (this bundle) versus agent staged-rollout planning (agent-production-operations).
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with production-excellence — SRE is shared. This bundle routes to it for reliability needs in delivery handoff; production-excellence routes to it for gate-entry SLO/error-budget evidence.
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with agent-production-operations — SRE is shared. This bundle uses it for delivery reliability; agent-production-operations uses it for runtime latency/cost budgets and incident handoff.
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with production-excellence — Security engineering is shared. Route by phase/domain: trust-boundary work in lifecycle phases (this bundle) versus security evidence for launch gates (production-excellence).
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with agent-production-operations — Security engineering is shared. Route by context: lifecycle-phase security needs (this bundle) versus agent authority contracts and disablement security (agent-production-operations).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with production-excellence — Platform engineering is shared. Route by context: lifecycle delivery handoff (this bundle) versus gate-entry platform evidence (production-excellence).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with agent-production-operations — Platform engineering is shared. Route by context: lifecycle delivery handoff (this bundle) versus agent sandboxing and tool-execution isolation (agent-production-operations).
  - privacy-engineering (`../../privacy-engineering/SKILL.md`) with agent-production-operations — Privacy engineering is shared. This bundle routes to it for PII, consent, retention, and data flows in any phase; agent-production-operations routes to it for trace scrubbing and user-data-access grants.
  - qa-methodology (`../../qa-methodology/SKILL.md`) with production-excellence — QA methodology is shared. Route by context: lifecycle phase 6 quality gates (this bundle) versus launch verification evidence (production-excellence).
  - verification-methodology (`../../verification-methodology/SKILL.md`) with production-excellence — Verification methodology is shared. This bundle uses it at every phase gate for evidence; production-excellence uses it for boundary labeling of launch evidence.
- **Eval suite:** evals/evals.json

### production-excellence

- Derivation: `manifest` — source: `bundles/production-excellence/manifest.yaml`
- **Purpose:** Route cross-domain production evidence (readiness, migration, recovery, capacity/cost, incident-learning) into a launch or operational decision — go, no-go, defer, exception, or escalation — with an accountable owner and a post-launch learning path.
- **Audience:** Service owners, SRE, release, platform, security, data, and QA teams coordinating a production change; agents assembling cross-domain production evidence into a defensible launch decision without re-deriving specialist runbooks.
- **Stages (ordered):**
  1. Cross-domain evidence assembly — production-readiness (`../../production-readiness/SKILL.md`), migration-engineering (`../../migration-engineering/SKILL.md`), resilience-and-recovery (`../../resilience-and-recovery/SKILL.md`), capacity-and-cost-engineering (`../../capacity-and-cost-engineering/SKILL.md`), incident-learning (`../../incident-learning/SKILL.md`)
  2. Gate evaluation — production-readiness (`../../production-readiness/SKILL.md`), site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`), verification-methodology (`../../verification-methodology/SKILL.md`), qa-methodology (`../../qa-methodology/SKILL.md`)
  3. Operational handoff — release-engineering (`../../release-engineering/SKILL.md`), platform-engineering (`../../platform-engineering/SKILL.md`), secure-software-engineering (`../../secure-software-engineering/SKILL.md`), data-engineering (`../../data-engineering/SKILL.md`)
  4. Post-launch learning — incident-learning (`../../incident-learning/SKILL.md`)
- **Included skills:**
  - production-readiness (`../../production-readiness/SKILL.md`)
  - migration-engineering (`../../migration-engineering/SKILL.md`)
  - resilience-and-recovery (`../../resilience-and-recovery/SKILL.md`)
  - capacity-and-cost-engineering (`../../capacity-and-cost-engineering/SKILL.md`)
  - incident-learning (`../../incident-learning/SKILL.md`)
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
  - release-engineering (`../../release-engineering/SKILL.md`)
  - platform-engineering (`../../platform-engineering/SKILL.md`)
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`)
  - data-engineering (`../../data-engineering/SKILL.md`)
  - qa-methodology (`../../qa-methodology/SKILL.md`)
  - verification-methodology (`../../verification-methodology/SKILL.md`)
- **Prerequisites:**
  - Readiness review evidence packet (risk-scaled, 11 categories) — via production-readiness (`../../production-readiness/SKILL.md`)
  - Migration plan, compatibility windows, and reconciliation evidence — via migration-engineering (`../../migration-engineering/SKILL.md`)
  - Resilience and recovery evidence (RTO/RPO, restore tests, game days) — via resilience-and-recovery (`../../resilience-and-recovery/SKILL.md`)
  - Capacity and cost model evidence (demand, scaling, unit cost) — via capacity-and-cost-engineering (`../../capacity-and-cost-engineering/SKILL.md`)
  - Incident history, causal hypotheses, and verified-closure records — via incident-learning (`../../incident-learning/SKILL.md`)
  - SLO and error-budget status — via site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`)
  - Release plan and rollback strategy — via release-engineering (`../../release-engineering/SKILL.md`)
- **Outputs:** gate-decision, operational-handoff-record, accountable-owner-record, evidence-gap-register
- **Handoffs:**
  - operational-handoff-record → incident-learning — Post-launch incidents (SLO degradations, unexpected failures, capacity breaches) route back to incident-learning with launch context; verified closure is required for follow-up items.
  - operational-handoff-record → product-lifecycle-learning — Expected outcomes recorded in the handoff (SLO targets, capacity assumptions, cost projections) feed expected-vs-observed comparison at the handoff review cadence.
  - gate-decision → agent-production-operations — Readiness outcomes (go/no-go/defer/exception) gate agent authority expansion when the runtime control plane is in use.
  - accountable-owner-record → service owner — Every gate outcome names an accountable owner and records the evidence supporting it; no gate passes on a bare checklist.
- **Conflicts:**
  - production-readiness (`../../production-readiness/SKILL.md`) with product-lifecycle — Both bundles include production-readiness. This bundle owns the launch gate decision; product-lifecycle routes phase 6 delivery handoff. Load the bundle matching the active decision context, never both at once.
  - production-readiness (`../../production-readiness/SKILL.md`) with agent-production-operations — agent-production-operations consumes readiness outcomes as authority- gating inputs. Readiness reviews and go/no-go/defer/exception outcomes are produced here and consumed there.
  - release-engineering (`../../release-engineering/SKILL.md`) with product-lifecycle — Release engineering is shared. Route by context: gate-entry release evidence (this bundle) versus lifecycle delivery handoff (product-lifecycle).
  - release-engineering (`../../release-engineering/SKILL.md`) with agent-production-operations — Release engineering is shared. Route by context: launch-gate release plans (this bundle) versus agent staged-rollout planning (agent-production-operations).
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with product-lifecycle — SRE is shared. This bundle routes to it for gate-entry SLO/error-budget evidence; product-lifecycle routes to it for delivery reliability.
  - site-reliability-engineering (`../../site-reliability-engineering/SKILL.md`) with agent-production-operations — SRE is shared. This bundle uses it for gate-entry reliability evidence; agent-production-operations uses it for runtime latency/cost budgets and incident handoff. Incident command stays with SRE in both cases.
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with product-lifecycle — Security engineering is shared. Route by context: security evidence for launch gates (this bundle) versus trust-boundary work in lifecycle phases (product-lifecycle).
  - secure-software-engineering (`../../secure-software-engineering/SKILL.md`) with agent-production-operations — Security engineering is shared. Route by context: launch-gate security evidence (this bundle) versus agent authority contracts and disablement security (agent-production-operations).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with product-lifecycle — Platform engineering is shared. Route by context: gate-entry platform evidence (this bundle) versus lifecycle delivery handoff (product-lifecycle).
  - platform-engineering (`../../platform-engineering/SKILL.md`) with agent-production-operations — Platform engineering is shared. Route by context: launch-gate platform evidence (this bundle) versus agent sandboxing and tool-execution isolation (agent-production-operations).
  - incident-learning (`../../incident-learning/SKILL.md`) with agent-production-operations — Both bundles consume incident-learning records. This bundle uses incident history as gate-entry evidence; agent-production-operations consumes verified-closure records as escalation and disablement inputs.
  - qa-methodology (`../../qa-methodology/SKILL.md`) with product-lifecycle — QA methodology is shared. Route by context: launch verification evidence (this bundle) versus lifecycle phase 6 quality gates (product-lifecycle).
  - verification-methodology (`../../verification-methodology/SKILL.md`) with product-lifecycle — Verification methodology is shared. This bundle uses it for boundary labeling of launch evidence; product-lifecycle uses it at every phase gate.
- **Eval suite:** evals/evals.json

### research-and-vault

- Derivation: `manifest` — source: `bundles/research-and-vault/manifest.yaml`
- **Purpose:** Chain web research, atomic extraction, and durable knowledge capture into a repeatable research-to-note workflow: capture source URLs and dates before interpretation, extract independent atomic claims before synthesis, and write the durable note only after source coverage and gaps are explicit.
- **Audience:** Agents and knowledge workers that repeatedly need the full research-to-note sequence (web research, atomic extraction, durable capture) and want a stable order of operations with a clear handoff artifact instead of ad-hoc steps; not for single lookups or single notes, which route to the domain skill directly.
- **Stages (ordered):**
  1. Web research — research-methodology (`../../research-methodology/SKILL.md`)
  2. Source capture — research-methodology (`../../research-methodology/SKILL.md`)
  3. Atomic extraction — research-methodology (`../../research-methodology/SKILL.md`)
  4. Synthesis and durable capture — research-methodology (`../../research-methodology/SKILL.md`)
- **Included skills:**
  - research-methodology (`../../research-methodology/SKILL.md`)
- **Prerequisites:**
  - A research question or topic and access to web-research tooling — via research-methodology (`../../research-methodology/SKILL.md`)
- **Outputs:** source-capture-record, atomic-claim-extract, durable-note
- **Handoffs:**
  - source-capture-record → research-methodology — Source URLs and dates are captured before interpretation and carried into extraction and synthesis; never interpret a source before its URL and date are recorded.
  - atomic-claim-extract → research-methodology — Independent atomic claims are extracted before synthesis so the durable note can trace each claim to its source.
  - durable-note → user — The finished durable note reports its output path and any unresolved gaps in source coverage.
- **Conflicts:** none declared
- **Eval suite:** evals/evals.json

### tailscale

- Derivation: `manifest` — source: `bundles/tailscale/manifest.yaml`
- **Purpose:** Operate a self-hosted Tailscale/Headscale ecosystem end to end: deploy and manage a Headscale control server, configure Tailscale clients, define ACL policy, manage node lifecycle, advertise subnet routes and exit nodes, operate DERP relays, and back up, restore, or migrate the control plane.
- **Audience:** Platform and homelab operators running self-hosted Tailscale/Headscale infrastructure; agents that need to know which sub-skill to load for deploy, policy, client, node, routing, DERP, or backup tasks in a WireGuard mesh.
- **Stages (ordered):**
  1. Deploy control server — headscale-deploy (`skills/headscale-deploy/SKILL.md`)
  2. Policy — tailnet-policy (`skills/tailnet-policy/SKILL.md`)
  3. Client connectivity — tailscale-client (`skills/tailscale-client/SKILL.md`)
  4. Node lifecycle — headscale-node-lifecycle (`skills/headscale-node-lifecycle/SKILL.md`)
  5. Routing — headscale-routing (`skills/headscale-routing/SKILL.md`)
  6. DERP relays — headscale-derp (`skills/headscale-derp/SKILL.md`)
  7. Backup and restore — headscale-backup (`skills/headscale-backup/SKILL.md`)
- **Included skills:**
  - headscale-deploy (`skills/headscale-deploy/SKILL.md`)
  - tailnet-policy (`skills/tailnet-policy/SKILL.md`)
  - tailscale-client (`skills/tailscale-client/SKILL.md`)
  - headscale-node-lifecycle (`skills/headscale-node-lifecycle/SKILL.md`)
  - headscale-routing (`skills/headscale-routing/SKILL.md`)
  - headscale-derp (`skills/headscale-derp/SKILL.md`)
  - headscale-backup (`skills/headscale-backup/SKILL.md`)
- **Prerequisites:**
  - A Linux host and install target for the Headscale control server — via headscale-deploy (`skills/headscale-deploy/SKILL.md`)
  - HEADSCALE_URL and HEADSCALE_API_KEY credentials — via headscale-deploy (`skills/headscale-deploy/SKILL.md`)
  - A running Headscale instance with the headscale CLI available — via headscale-deploy (`skills/headscale-deploy/SKILL.md`)
- **Outputs:** headscale-server, tailnet-policy, registered-nodes, routed-networks, derp-map, backup-archive, restored-instance
- **Handoffs:**
  - headscale-server → tailnet-policy — A running control server is required before ACL policy is applied and before the tailnet opens to other users.
  - headscale-server → headscale-node-lifecycle — Nodes are registered, approved, tagged, and decommissioned against the running control server.
  - headscale-server → headscale-backup — Backups (sqlite + config + policy + certs) run regularly against the production control server; restore re-creates the instance from the archive.
  - backup-archive → user — The backup archive is the disaster-recovery artifact for the tailnet; keep it off the control server host.
- **Conflicts:** none declared
- **Eval suite:** evals/evals.json

### workflow-architect

- Derivation: `manifest` — source: `bundles/workflow-architect/manifest.yaml`
- **Purpose:** Discover a user's actual workflow through active interrogation or passive observation, then generate a tailored Agent Skills bundle that encodes it as loadable skills with trigger conditions, a decision map, a manifest, and an optional kanban board — turning a session that feels aimless into structure.
- **Audience:** Users and agents that want to formalize how they work — understand their own process, encode it as trigger-driven skills they can reuse, or share it with collaborators. Not for single factual questions or tasks already owned by a more specific skill.
- **Stages (ordered):**
  1. Active interview — interviewer (`skills/interviewer/SKILL.md`)
  2. Passive observation — observer (`skills/observer/SKILL.md`)
  3. Synthesis and generation — bundle-builder (`skills/bundle-builder/SKILL.md`)
- **Included skills:**
  - interviewer (`skills/interviewer/SKILL.md`)
  - observer (`skills/observer/SKILL.md`)
  - bundle-builder (`skills/bundle-builder/SKILL.md`)
- **Prerequisites:**
  - Session context with enough turns for observation (or user answers for interview) — via observer (`skills/observer/SKILL.md`)
  - Convergence on workflow phases, branching signals, and tool preferences — via interviewer (`skills/interviewer/SKILL.md`)
- **Outputs:** skills-bundle, decision-map, manifest, kanban-board
- **Handoffs:**
  - skills-bundle → user — The generated bundle is written to the host skill directory and registered so the umbrella and sub-skills load on trigger phrases.
  - decision-map → user — The Mermaid decision map visualizes workflow phases and branching signals for review and sharing.
  - manifest → user — The generated manifest maps skill names to trigger conditions, entry points, and transition signals for the bundle.
  - kanban-board → user — The kanban board is included only when the workflow follows a predictable linear path where WIP limits and lane transitions add value.
- **Conflicts:** none declared
- **Eval suite:** evals/evals.json