agentsclimarketplace

Sheepdog review

Skill filipores/sheepdog/skills/sheepdog-review

Review a diff or target for agentic architecture drift and boundary-law violations. Use for sheepdog review, architecture drift review, god orchestrator review, layer-boundary review, or before merging AI-generated code. Report only; do not edit.From its SKILL.md

Install
npx -y skills add filipores/sheepdog --skill sheepdog-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.6 KB, 320 tokens by cl100k_base, as published. Nobody here has run it

Sheepdog Review

Review the current diff or named target for misplaced responsibility. Report only. Do not fix.

Laws

Use Sheepdog laws 1-8: entrypoint, orchestrator, decision-owner, side-effect, persistence, pattern-reuse, dependency-direction, god-gravity.

Method

  1. Read repo guidance first when present: AGENTS.md, ARCHITECTURE.md, ADRs, architecture/domain docs relevant to the target.
  2. Inspect changed files and nearby existing patterns.
  3. Flag only concrete boundary violations with file/line evidence.
  4. Size is only a signal. Do not block merely because a file is long.

Format

One finding per line:

<file>:L<line>: law<n> <tag>: <stray sheep>. wrong field: <current>. home: <right field>. move: <smallest safe move>.

Tags:

  • entrypoint: handler owns non-routing work
  • orchestrator: coordinator owns rules/details/effects
  • decision: policy/state logic in wrong place
  • side-effect: external effect bypasses adapter/guard
  • persistence: direct storage access from wrong layer
  • reuse: parallel implementation instead of existing pattern
  • dependency: wrong-way import or cycle
  • god-gravity: new responsibility added to god file

End with:

verdict: BLOCKED (<N> boundary leaks)

or:

All sheep in field. Ship.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most review quality skills give in 320 tokens

Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07

  • Ask questions one at a timein 81 of 1048, across 64 files
  • Provide a recommended answer for each questionin 73 of 1048, across 50 files
  • Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
  • Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
  • Interview the user relentlessly about the planin 38 of 1048, across 13 files
  • Order findings by severityin 31 of 1048
  • Resolve each branch of the decision treein 27 of 1048, across 5 files
  • Run a grilling sessionin 26 of 1048, across 5 files
  • Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
  • Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
  • Create documentation files lazilyin 24 of 1048, across 5 files
  • Assign severity to every findingin 24 of 1048

Said here and by no other author read

  • read repo guidance before inspecting the target
  • inspect changed files and nearby existing patterns
  • flag only concrete boundary violations with file and line evidence
  • apply sheepdog laws one through eight
  • state the smallest safe move to fix the violation
  • ignore file length as a sole blocking reason

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,149. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.