agentleFS
sponge-b0bsponge-b0b/Polaris/.agents/skills/to-doc/SKILL.md

to-doc

Skill4 starsChanged 4 months agoApache-2.0View on GitHub

Create new Living Entity Wiki-classified non-ADR documents under docs/ using the classification, naming, placement, and Wiki rules defined by the repository schema.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

32 rules. About 0 are common, 16 are distinctive, and 16 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈16

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 16

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill only for a new non-ADR document inside the Living Entity Wiki document universe.

    line 33
  2. Very specific:

    If the content warrants an ADR, use $to-adr-doc.

    line 35
  3. Very specific:

    If the document already exists, use $classify-doc.

    line 36
  4. Very specific:

    If the requested path is matched by docs/.wikiignore, it is outside this skill's ownership.

    line 37
  5. Distinctive:

    Do not recreate an existing document merely to bypass reclassification or reference migration.

    line 38
  6. Very specific:

    If current vs proposed is genuinely unclear, follow the schema fail-safe and use proposed.

    line 65
  7. Very specific:

    Read docs/.wikiignore when present and apply only the exclusion grammar defined by wiki/_schema.md.

    line 69
  8. Distinctive:

    If the requested path is explicitly excluded:

    line 71
  9. Distinctive:

    do not classify it;

    line 73
  10. Distinctive:

    do not apply Living Entity Wiki naming rules.

    line 74
  11. Very specific:

    do not invoke $wiki-sync solely because it was created.

    line 75
  12. Very specific:

    Do not infer exclusions for unfamiliar directories and do not invent new .wikiignore entries merely to make a requested path legal.

    line 78
  13. Very specific:

    When an active wiki/index.md exists:

    line 84
  14. Distinctive:

    use the owning Entity ID when the document primarily belongs to one entity.

    line 86
  15. Very specific:

    use platform- only for genuinely cross-cutting documents.

    line 87
  16. Very specific:

    do not entity/platform-prefix process documents.

    line 88
  17. Very specific:

    Use wiki/index.md as the source of active Entity IDs.

    line 90
  18. Distinctive:

    Do not invent new IDs.

    line 92
  19. Distinctive:

    If no active entity registry exists yet, use:

    line 96
  20. Distinctive:

    Do not invent entity prefixes before bootstrap.

    line 102
  21. Distinctive:

    Use the canonical folder for its class:

    line 108
  22. Distinctive:

    Create only the required directory if missing.

    line 118
  23. Distinctive:

    Write the file directly at its canonical destination.

    line 120
  24. Distinctive:

    create it temporarily at the repository root.

    line 124
  25. Very specific:

    leave it loose under docs/.

    line 125
  26. Distinctive:

    If the Living Entity Wiki has not been bootstrapped, no synchronization is required.

    line 130
  27. Distinctive:

    Do not duplicate that evaluation here.

    line 140
  28. Distinctive:

    Do not turn the index into a complete document inventory.

    line 144
  29. Very specific:

    If document creation also causes wiki/index changes, keep those changes consistent with the caller's commit strategy and $wiki-sync's …

    line 150
  30. Very specific:

    Do not create a separate commit merely because $to-doc or $wiki-sync ran.

    line 152
  31. Very specific:

    create or manage documentation matched by docs/.wikiignore.

    line 169
  32. Very specific:

    create or manage ADRs — use $to-adr-doc.

    line 171

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

To Doc

$to-doc owns creation-time placement and naming of new Living Entity Wiki-classified non-ADR documents under docs/.

Its responsibility is:

content
  ↓
classify
  ↓
name
  ↓
write
  ↓
sync when required

Classification, naming, and the docs/.wikiignore exclusion boundary are owned by wiki/_schema.md.

For an existing classified document, use $classify-doc.

For an ADR, use $to-adr-doc.

1. Confirm Scope

Use this skill only for a new non-ADR document inside the Living Entity Wiki document universe.

  • If the content warrants an ADR, use $to-adr-doc.
  • If the document already exists, use $classify-doc.
  • If the requested path is matched by docs/.wikiignore, it is outside this skill's ownership; use the workflow or repository owner responsible for that documentation tree.
  • Do not recreate an existing document merely to bypass reclassification or reference migration.

2. Classify the Document

Apply the non-ADR classification rules in wiki/_schema.md.

Possible project-owned classes are:

current
proposed
research
reference
process

Classification comes from folder placement.

Do not add:

doc_class:
Doc-Class:

to the document.

If current vs proposed is genuinely unclear, follow the schema fail-safe and use proposed.

3. Respect the Wiki Exclusion Boundary

Read docs/.wikiignore when present and apply only the exclusion grammar defined by wiki/_schema.md.

If the requested path is explicitly excluded:

  • do not classify it;
  • do not apply Living Entity Wiki naming rules;
  • do not invoke $wiki-sync solely because it was created;
  • route creation to the workflow or repository owner responsible for that documentation tree.

Do not infer exclusions for unfamiliar directories and do not invent new .wikiignore entries merely to make a requested path legal.

4. Name the Document

Apply the naming and cross-cutting-document rules in wiki/_schema.md.

When an active wiki/index.md exists:

  • use the owning Entity ID when the document primarily belongs to one entity;
  • use platform- only for genuinely cross-cutting documents;
  • do not entity/platform-prefix process documents.

Use wiki/index.md as the source of active Entity IDs.

Do not invent new IDs.

Before Wiki Bootstrap

If no active entity registry exists yet, use:

<slug>.md

Do not invent entity prefixes before bootstrap.

Later normalization belongs to $classify-doc.

5. Place and Write the Document

Use the canonical folder for its class:

current   → docs/current/
proposed  → docs/proposed/
research  → docs/research/
reference → docs/reference/
process   → docs/process/

Create only the required directory if missing.

Write the file directly at its canonical destination.

Do not:

  • create it temporarily at the repository root;
  • leave it loose under docs/;
  • duplicate classification or entity-registry metadata inside it.

6. Living Entity Wiki Follow-Through

If the Living Entity Wiki has not been bootstrapped, no synchronization is required.

Otherwise:

  • new current documents → invoke $wiki-sync;
  • new proposed documents → invoke $wiki-sync;
  • new research, reference, or process documents → do not invoke $wiki-sync solely because they were created.

Let $wiki-sync determine whether a current or proposed document contributes durable entity knowledge or exposes [source-conflict].

Do not duplicate that evaluation here.

For genuinely cross-cutting platform- documents, maintain any required discovery entry in wiki/index.md according to wiki/_schema.md.

Do not turn the index into a complete document inventory.

7. Commit Ownership

$to-doc does not require its own commit when called inside a parent workflow.

If document creation also causes wiki/index changes, keep those changes consistent with the caller's commit strategy and $wiki-sync's semantic logging rules.

Do not create a separate commit merely because $to-doc or $wiki-sync ran.

8. Report

Report:

  • created path;
  • classification;
  • primary entity or platform attribution where applicable;
  • whether $wiki-sync ran;
  • whether wiki/index.md changed;
  • any [source-conflict] or unresolved entity-boundary issue.

Out of Scope

$to-doc does not:

  • create or manage documentation matched by docs/.wikiignore;
  • modify or reclassify existing documents — use $classify-doc;
  • create or manage ADRs — use $to-adr-doc;
  • own classification, naming, or exclusion policy — see wiki/_schema.md;
  • decide entity topology — use $wiki-sync;
  • resolve [source-conflict];
  • maintain linked_docs;
  • perform a full wiki audit — use the $wiki-lint skill.

Its job is to ensure a new classified non-ADR document starts in the correct place, with the correct name, and the correct lifecycle follow-through.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.