agentleFS
s977043s977043/river-review/skills/midstream/agent-skill-bridge/SKILL.md

Agent Skill Bridge Review

Skill7 starsChanged 5 months agoMITView on GitHub

Review changes to the Agent Skills import/export bridge for path safety, round-trip fidelity, and validation correctness.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Test coverage: every newly exported function from the module must have at least one node:test case.

    line 38
  2. Distinctive:

    Do not critique the RR skill schema itself.

    line 42
  3. Distinctive:

    Do not flag auto-fill behaviour as missing validation when the converted metadata passes the strict RR schema.

    line 43
  4. Distinctive:

    If test fixtures are added without corresponding logic changes, that is expected (test-only commits).

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Pattern declaration

Primary pattern: Reviewer Secondary patterns: Inversion Why: ブリッジモジュールのパス安全性・往復忠実性・バリデーション正当性をチェックリスト型で評価するが、対象ファイル外の変更では実行不要

Guidance

  • Path traversal: assertSafePath must reject ids containing .., /, or NUL bytes. Any new file-write path must route through this function or sanitizeSkillId.
  • Round-trip fidelity: export followed by re-import must preserve name, description, and body content. Verify that new fields added to conversion logic do not break this property.
  • Auto-fill correctness: convertAgentSkillToRR must not overwrite explicitly provided values (id, category, phase, applyTo). Defaults apply only when the field is absent.
  • Validation gate: both strict and loose validation paths must be exercised. The errors array must capture invalid skills without aborting the entire import batch.
  • Test coverage: every newly exported function from the module must have at least one node:test case.

Non-goals

  • Do not critique the RR skill schema itself; focus on bridge logic only.
  • Do not flag auto-fill behaviour as missing validation when the converted metadata passes the strict RR schema.

Pre-execution Gate / 実行前ゲート

このスキルは以下の条件がすべて満たされない限りNO_REVIEWを返す。

  • 差分にAgent Skillブリッジ関連ファイル(agent-skill-bridge.mjs, validate-agent-skills.mjs, agent-skill-loose.schema.json等)の変更が含まれている
  • 差分がコメント・JSDoc・空白のみの変更ではない
  • inputContextにdiffが含まれている

ゲート不成立時の出力: NO_REVIEW: agent-skill-bridge — Agent Skillブリッジ関連の実質的なコード変更が検出されない

False-positive guards

  • If test fixtures are added without corresponding logic changes, that is expected (test-only commits).

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.