AI & Agentslow risk
atm10-source-of-truth-check
Explicit activation required: do not invoke or load this skill from an implicit match; wait for explicit user or operator invocation or a source-authorized parent-route selection. Apply the aoa-source-of-truth-check workflow inside an atm10-* repository using repo-relative public-surface roles, document maps, canonical-file patterns, entrypoint trimming, and local review posture. Use when contributors need a thin project overlay to identify authoritative ATM10 docs or separate active, archived, generated, local-only, and runtime-adjacent surfaces. Do not use when the task is broader policy design, purely code-local, runtime authority, or better handled by the base skill without local adaptation.
8Dionysus/aoa-skills·.agents/skills/atm10-source-of-truth-check/SKILL.md
37/ 100おすすめ度
この Skill を導入
coding agent を選び、プロジェクト用または個人用コマンドをコピーします。
プロジェクトに導入.agents/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a codex -y個人環境に導入~/.agents/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a codex -g -yプロジェクトに導入.claude/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a claude-code -y個人環境に導入~/.claude/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a claude-code -g -yプロジェクトに導入.agents/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a github-copilot -y個人環境に導入~/.copilot/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a github-copilot -g -yプロジェクトに導入.agents/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a cursor -y個人環境に導入~/.cursor/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a cursor -g -yプロジェクトに導入.agents/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a gemini-cli -y個人環境に導入~/.gemini/skills/atm10-source-of-truth-check
npx skills add https://github.com/8Dionysus/aoa-skills/tree/ec79a94b8c1767fbd6281326767b6cb83a1d6673/.agents/skills/atm10-source-of-truth-check -a gemini-cli -g -yNative Gemini CLI
gemini skills install https://github.com/8Dionysus/aoa-skills.git --scope workspace --path .agents/skills/atm10-source-of-truth-check⚠ インストールには open-source skills CLI を使用します。実行前にソースと権限を確認してください。
Skill の指示
GitHub で元ファイルを表示 ↗# atm10-source-of-truth-check ## Intent Use this skill to adapt `aoa-source-of-truth-check` to an `atm10-*` repository when the base workflow is right but the local public-doc and repo-surface map needs repo-relative detail. ## Trigger boundary Use this skill when: - the base `aoa-source-of-truth-check` workflow is already correct, but an `atm10-*` repo needs local canonical-file patterns, repo-relative docs, or doc review rules - contributors need a thin overlay that maps repo-relative docs such as `README.md`, `docs/ARCHITECTURE.md`, or `docs/[canonical-guide].md` - confusion exists between overview docs and authoritative files inside one local repo - active commands, archived commands, public status, support-profile claims, runtime baseline notes, release posture, or local-only planning surfaces are being mixed - a top-level ATM10 entrypoint is accumulating runbook-scale detail that should move to a canonical home - a public-safe local route must distinguish tracked docs from ignored maintainer scratch, internal chronology, private logs, or tool-local config - the family review doc and bundle-local checklist still need to stay aligned Do not use this skill when: - the main need is broader policy design rather than local document authority mapping - the task is purely code-local and has no meaningful docs or guidance ambiguity - the work would introduce new upstream technique meaning instead of thin local adaptation - the main need is recording rationale for a decision rather than clarifying authority; use `aoa-adr-write` - the authoritative files are already clear and the remaining work is an ordinary bounded change; use `atm10-change-protocol` or the base `aoa-change-protocol` - the question belongs to ATM10 runtime behavior, perception truth, model-host selection, operator automation, or private local state rather than document authority ## Inputs - repo-relative docs or guidance surfaces - local canonical-file candidates - local review rules for doc changes - contributor confusion points - ATM10 public-surface roles from `docs/SOURCE_OF_TRUTH.md` - affected active, archived, generated/export, local-only, internal, or runtime-adjacent surfaces - candidate validation path such as public-doc hardening tests, nested AGENTS checks, or a no-run review note - base skill reference ## Outputs - local source-of-truth map - active/current versus archived, generated/export, local-only, internal, or runtime-adjacent placement map - bounded clarification note - repo-relative canonical-file pattern - lightweight entrypoint guidance when a summary doc should route outward instead of duplicating commands or counters - pointer to the family review surface - verification summary for the local docs surface ## Procedure 1. start from `aoa-source-of-truth-check` instead of inventing a family-specific docs doctrine 2. read the ATM10 root route card and `docs/SOURCE_OF_TRUTH.md`; if a touched path has a nested `AGENTS.md`, read that before judging local authority 3. name the repo-relative docs, source/config/schema surfaces, generated/export companions, run artifacts, or local-only surfaces involved in the ambiguity 4. map each local concern to the smallest authoritative home, such as `MANIFEST.md` for public current state, `ROADMAP.md` for direction, `docs/RUNBOOK.md` for active runnable commands, `docs/ARCHIVED_TRACKS.md` for archived/recoverable command references, `docs/PRODUCT_EDGE_POSTURE.md` for support/test-tier claims, `docs/QWEN3_MODEL_STACK.md` for model/runtime baseline posture, or `docs/SOURCE_OF_TRUTH.md` for document-role rules 5. keep `README.md` and other entrypoints short and link-driven once the canonical home exists 6. separate tracked public docs from ignored local-only planning, internal chronology, private logs, host-specific paths, tokens, and tool-local config 7. keep the adaptation bounded to the local repo surface under review 8. make explicit what still depends on downstream human review, unpublished local policy, or ATM10-owned runtime evidence 9. verify that the clarification reduces ambiguity without making `aoa-skills` the authority for ATM10 behavior ## Contracts - preserve the base skill meaning - keep local file maps repo-relative and explicit - surface local authority and review posture without hiding it - keep the overlay public-safe and reviewable - keep ATM10 source-of-truth claims inside ATM10-owned public docs and route cards - keep active runbook content separate from archived or recoverable references - keep generated, exported, installed, and run-artifact surfaces weaker than the authored files they summarize - keep public support-profile claims no broader than the docs and test tiers that actually validate them - keep private workstation paths, local model paths, private logs, hostnames, tokens, screenshots with sensitive details, and internal scratch out of public examples ## Risks and anti-patterns - inventing a broader docs governance framework inside a thin overlay - using family labels without reducing local ambiguity - hiding local review rules in prose that looks canonical - silently replacing the base skill with project doctrine - treating `README.md`, generated outputs, run artifacts, or installed skills as the source of truth when ATM10 already has canonical docs - mixing active commands with archived/recoverable references - spreading release cadence, supported profiles, or test-tier claims across several docs when one canonical surface owns them - turning ignored local-only planning or internal chronology into public truth - using this overlay to make runtime, service, model, operator automation, or perception decisions that belong in ATM10-owned implementation and docs ## Verification - confirm the base skill is still the right workflow - confirm authoritative repo-relative files are named explicitly - confirm local review posture is visible rather than implied - confirm the adaptation reduces ambiguity without widening scope - confirm active, archived, generated/export, local-only, internal, and runtime-adjacent surfaces are not collapsed into one role - confirm entrypoint docs stay short and route to canonical homes where those exist - confirm public-safe boundaries and private/local-only exclusions are preserved - confirm any support-profile, release, or runtime-baseline wording points to the owning ATM10 surface - confirm the family review doc and bundle-local checklist stay aligned ## Technique traceability Manifest-backed techniques: - AOA-T-0013 from `8Dionysus/aoa-techniques` at `cd276f040d55d490bd015b8698c7a5d594b9f875` using path `techniques/instruction/instruction-surface/single-source-rule-distribution/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation - AOA-T-0002 from `8Dionysus/aoa-techniques` at `cd276f040d55d490bd015b8698c7a5d594b9f875` using path `techniques/instruction/docs-boundary/source-of-truth-layout/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation - AOA-T-0009 from `8Dionysus/aoa-techniques` at `cd276f040d55d490bd015b8698c7a5d594b9f875` using path `techniques/instruction/docs-boundary/lightweight-status-snapshot/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation ## Adaptation points - local doc hierarchies - repo-relative canonical-file patterns - local review rules for doc changes - repository-specific authority examples - lightweight entrypoint rules for `README.md`, `MANIFEST.md`, and other summary docs - active versus archived command placement - public-safe exclusions for internal scratch, private logs, host-specific paths, and tool-local config - family review doc and bundle-local review checklist