aoa-source-of-truth-check
Intent
Use this skill to clarify which files are authoritative for status, architecture, run instructions, policy, and change guidance.
Trigger boundary
Use this skill when:
- a repository has several docs that may overlap or conflict
- contributors may not know which file to trust first
- a change touches docs, process, or operational guidance and the question is which file is authoritative
- confusion exists between overview docs and authoritative docs
- one authoritative source must stay aligned across multiple downstream consumer surfaces
- top-level status docs such as
README or MANIFEST are accumulating status/history that should live in canonical homes instead
- the repository already has canonical detail surfaces and the summary docs should stay short, navigable, and link-driven
Do not use this skill when:
- the repository is tiny and has no meaningful source-of-truth ambiguity
- the task is purely code-local with no documentation or policy impact
- the authoritative files are already clear and the main need is recording rationale for a decision; use
aoa-adr-write
- the main problem is deciding whether logic belongs in the core or at the edge; use
aoa-core-logic-boundary first
- the main problem is broader policy design rather than document authority or ownership
- the task is only about building or maintaining a derived docs surface; that belongs in a separate review-surface workflow
Inputs
- repository docs surface
- target area of ambiguity or overlap
- known canonical files if any
- current contributor confusion points
Outputs
- clearer source-of-truth map
- fan-out map when one source feeds multiple downstream consumers
- note of overlaps or conflicts
- proposed or implemented document role clarification
- lightweight snapshot guidance for entrypoint docs when canonical homes already exist
- verification summary
Procedure
- identify the main docs or guidance files involved in the target area
- determine which file should be authoritative for each concern
- note any overlap, contradiction, or role ambiguity
- if one source feeds multiple consumers, name each consumer and refresh them from the same source
- if top-level status docs are bloating, trim them into short snapshots and route detail to canonical homes
- clarify or propose clarifying document ownership and purpose
- keep the change bounded to the guidance surface under review
- verify that the result reduces ambiguity for future changes
Contracts
- authoritative sources should be visible and named explicitly
- overview documents should not silently replace canonical ones
- lightweight entrypoint docs should link outward instead of duplicating chronology or changing counters
- role separation should reduce confusion, not create extra ceremony
- the resulting guidance should be understandable to another human or agent
Risks and anti-patterns
- over-formalizing a tiny docs surface
- creating many labels without reducing ambiguity
- moving truth across files without clearly signaling the change
- letting summaries masquerade as canonical instructions
- trimming top-level docs too aggressively before canonical homes are actually available
- widening the skill into generic docs hygiene or derived surface maintenance
Verification
- confirm the main source-of-truth ambiguity was reduced
- confirm authoritative files are named explicitly
- confirm overlaps or conflicts were surfaced rather than hidden
- confirm summary docs stay short and route detail to canonical homes where those already exist
- confirm the result helps future contributors orient faster
Technique traceability
Manifest-backed techniques:
- AOA-T-0013 from
8Dionysus/aoa-techniques at 5c6f0496edc3c2e74590baa35627c85fe58ef765 using path techniques/docs/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 5c6f0496edc3c2e74590baa35627c85fe58ef765 using path techniques/docs/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 5c6f0496edc3c2e74590baa35627c85fe58ef765 using path techniques/docs/lightweight-status-snapshot/TECHNIQUE.md and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation
Adaptation points
Future project overlays may add:
- local doc hierarchies
- mixed ADR, architecture, and operations doc maps that split authority by concern
- preferred canonical-file patterns
- local review rules for doc changes
- repository-specific examples of authoritative surfaces
- lightweight snapshot rules for README or MANIFEST surfaces
- rules for keeping entrypoint docs short once deeper canonical homes already exist
1---2name: aoa-source-of-truth-check3description: Clarify which files are authoritative for repository guidance, operational instructions, architecture, and status, and keep entrypoint docs short and link-driven once canonical homes exist. Use when docs overlap, conflict, or confuse contributors about which file to trust. Do not use for purely code-local tasks or when authoritative files are already clear and the main need is decision rationale.4license: Apache-2.05---6
7# aoa-source-of-truth-check
8
9## Intent
10Use this skill to clarify which files are authoritative for status, architecture, run instructions, policy, and change guidance.
11
12## Trigger boundary
13Use this skill when:
14- a repository has several docs that may overlap or conflict
15- contributors may not know which file to trust first
16- a change touches docs, process, or operational guidance and the question is which file is authoritative
17- confusion exists between overview docs and authoritative docs
18- one authoritative source must stay aligned across multiple downstream consumer surfaces
19- top-level status docs such as `README` or `MANIFEST` are accumulating status/history that should live in canonical homes instead
20- the repository already has canonical detail surfaces and the summary docs should stay short, navigable, and link-driven
21
22Do not use this skill when:
23- the repository is tiny and has no meaningful source-of-truth ambiguity
24- the task is purely code-local with no documentation or policy impact
25- the authoritative files are already clear and the main need is recording rationale for a decision; use `aoa-adr-write`
26- the main problem is deciding whether logic belongs in the core or at the edge; use `aoa-core-logic-boundary` first
27- the main problem is broader policy design rather than document authority or ownership
28- the task is only about building or maintaining a derived docs surface; that belongs in a separate review-surface workflow
29
30## Inputs
31- repository docs surface
32- target area of ambiguity or overlap
33- known canonical files if any
34- current contributor confusion points
35
36## Outputs
37- clearer source-of-truth map
38- fan-out map when one source feeds multiple downstream consumers
39- note of overlaps or conflicts
40- proposed or implemented document role clarification
41- lightweight snapshot guidance for entrypoint docs when canonical homes already exist
42- verification summary
43
44## Procedure
451. identify the main docs or guidance files involved in the target area
462. determine which file should be authoritative for each concern
473. note any overlap, contradiction, or role ambiguity
484. if one source feeds multiple consumers, name each consumer and refresh them from the same source
495. if top-level status docs are bloating, trim them into short snapshots and route detail to canonical homes
506. clarify or propose clarifying document ownership and purpose
517. keep the change bounded to the guidance surface under review
528. verify that the result reduces ambiguity for future changes
53
54## Contracts
55- authoritative sources should be visible and named explicitly
56- overview documents should not silently replace canonical ones
57- lightweight entrypoint docs should link outward instead of duplicating chronology or changing counters
58- role separation should reduce confusion, not create extra ceremony
59- the resulting guidance should be understandable to another human or agent
60
61## Risks and anti-patterns
62- over-formalizing a tiny docs surface
63- creating many labels without reducing ambiguity
64- moving truth across files without clearly signaling the change
65- letting summaries masquerade as canonical instructions
66- trimming top-level docs too aggressively before canonical homes are actually available
67- widening the skill into generic docs hygiene or derived surface maintenance
68
69## Verification
70- confirm the main source-of-truth ambiguity was reduced
71- confirm authoritative files are named explicitly
72- confirm overlaps or conflicts were surfaced rather than hidden
73- confirm summary docs stay short and route detail to canonical homes where those already exist
74- confirm the result helps future contributors orient faster
75
76## Technique traceability
77Manifest-backed techniques:
78- AOA-T-0013 from `8Dionysus/aoa-techniques` at `5c6f0496edc3c2e74590baa35627c85fe58ef765` using path `techniques/docs/single-source-rule-distribution/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation
79- AOA-T-0002 from `8Dionysus/aoa-techniques` at `5c6f0496edc3c2e74590baa35627c85fe58ef765` using path `techniques/docs/source-of-truth-layout/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation
80- AOA-T-0009 from `8Dionysus/aoa-techniques` at `5c6f0496edc3c2e74590baa35627c85fe58ef765` using path `techniques/docs/lightweight-status-snapshot/TECHNIQUE.md` and sections: Intent, When to use, Inputs, Outputs, Core procedure, Contracts, Risks, Validation
81
82## Adaptation points
83Future project overlays may add:
84- local doc hierarchies
85- mixed ADR, architecture, and operations doc maps that split authority by concern
86- preferred canonical-file patterns
87- local review rules for doc changes
88- repository-specific examples of authoritative surfaces
89- lightweight snapshot rules for README or MANIFEST surfaces
90- rules for keeping entrypoint docs short once deeper canonical homes already exist