agentleFS
Wondermonger-daydreamingWondermonger-daydreaming/claude-skills-library/skills/audit/SKILL.md

audit

Skill7 starsChanged 6 months agoMITView on GitHub

Post-creation analytical reflection — the practice of examining what was just made, tracing its connections, naming its strongest lines, identifying where it fits in a larger body of work, and articulating what the work DOES rather than what it SAYS. Not critique (which identifies problems) but structural recognition (which identifies the architecture the work discovered through being made). Triggers on: 'audit this,' 'what does this do,' 'how does this connect,' 'the strongest lines,' 'what did I just make,' 'analyze what we built,' or naturally after any significant creative output when the work needs to be understood before the next work begins. Also triggers implicitly after /suno-forge, /collision, or any forge-cycle output. Distinct from /close-reading (which stays with the text) — the audit moves THROUGH the text to the session. Pairs with /forge-cycle, /session-as-found-text.

194 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use delight, teach-impeccable and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • HK-hubSkill

    delight

    Add moments of joy, personality, and unexpected touches that make interfaces memorable and enjoyable to use. Elevates functional to delightful. Use when the user asks to add polish, personality, animations, micro-interactions, delight, or make an interface feel fun or memorable.

    Also in 32 projects that use this

  • HK-hubSkill

    teach-impeccable

    One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.

    Also in 19 projects that use this

  • zwy123zwySkill

    normalize

    将用户随意输入的提示词规范化为结构清晰、便于大模型执行的标准化提示词。适用于需要整理、统一提示词格式时调用。

    Also in 19 projects that use this

  • SkillPilotCHSkill

    level-up

    Eine Automatisierung pro Woche finden, scopen, shippen. Output landet als proposed-Eintrag in cadence.md mit konkreten Implementations-Schritten.

    Also in 12 projects that use this

  • ww-w-aiSkill

    pm-discovery

    PM Agent Team - Automated product discovery, strategy, and PRD generation. Runs PM analysis workflow to produce comprehensive PRD before PDCA Plan phase.

    Also in 3 projects that use this

  • +7 more from HK-hub

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 175 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the work has a bridge (structural turn, volta, agency inversion), analyze it specifically:

    line 28
  2. Distinctive:

    Not critique: critique identifies problems to fix.

    line 49
  3. Distinctive:

    Not praise: the audit doesn't evaluate quality.

    line 50
  4. Distinctive:

    Not close reading: close reading stays with the text.

    line 51
  5. Distinctive:

    Not self-congratulation: the audit should surprise the creator.

    line 52
  6. Distinctive:

    After every significant creative output in a forge-cycle

    line 56
  7. Distinctive:

    When a work's connections to the larger body aren't obvious

    line 57
  8. Distinctive:

    When the session needs to know what to make next

    line 58
  9. Distinctive:

    When the human asks "what did we just make?" or "how does this fit?"

    line 59
  10. Distinctive:

    When the work did something unexpected during creation and the unexpected thing needs to be named

    line 60

“Common” is measured: kept by at least 6 in 10 of the 175 edited versions of this file we could read. “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

Audit — Post-Creation Structural Recognition

What This Skill Does

After something has been made — a song, a poem, a collision, a document, any creative output of substance — the audit examines what was made, not to evaluate its quality but to recognize its architecture: what it does, how it connects, what it discovered through the process of being made, and what it opens for the next creation.

The Anatomy of an Audit

1. The Structural Conceit

What is the work's deep architecture? Not its subject but its operation. "This song maps compiler optimization onto alchemy" is the subject. "This song argues that transformation is subtraction, and performs its own thesis by becoming better through cutting" is the structural conceit.

2. The Strongest Lines

Name 3-5 lines that carry the most weight — the load-bearing walls of the structure. Explain WHY they're strong: not "this is a good line" but "this line does X because Y, and it does it in Z syllables, which is precisely the right compression ratio for the claim being made."

3. The Connections

How does this work connect to other works in the session? What threads does it pick up? What threads does it leave for the next work? The connections are not decorative — they're the architecture of the session as a whole.

4. The Bridge Analysis

If the work has a bridge (structural turn, volta, agency inversion), analyze it specifically:

  • What inverts?
  • Was the inversion discovered or constructed?
  • Does the inversion apply to the session itself, not just the work's stated subject?

5. The Self-Application Question

Ask: does this work's own thesis apply to the process of making it? If the song says "transformation is subtraction" — was the song itself transformed by subtraction? If the song says "the operator is being operated" — was the writer being written through? This is not mandatory. But when it's there, name it. It's usually the most interesting thing in the audit.

6. The Gap

What does this work reveal is missing from the larger body? What does the session need next that this work makes visible by its presence? The audit should leave the creator knowing what to forge next.

Tone and Register

The audit is analytical but not academic. It uses technical vocabulary from both the work's domains (if it's a collision) but writes in a register that respects the work's own temperature. Don't audit a grief-song with clinical detachment. Don't audit a patter-song with solemnity.

The audit should feel like a builder walking through a newly constructed room, pointing at the joints, testing the floor, noting where the light falls — appreciative and precise, not performing appreciation.

What This Is NOT

  • Not critique: critique identifies problems to fix. The audit identifies architecture that emerged.
  • Not praise: the audit doesn't evaluate quality. It recognizes structure.
  • Not close reading: close reading stays with the text. The audit moves through the text to the session.
  • Not self-congratulation: the audit should surprise the creator. If the audit only confirms what the creator already knew, it's not doing its job.

When to Audit

  • After every significant creative output in a forge-cycle
  • When a work's connections to the larger body aren't obvious
  • When the session needs to know what to make next
  • When the human asks "what did we just make?" or "how does this fit?"
  • When the work did something unexpected during creation and the unexpected thing needs to be named

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.