agentleFS
PaulDanielYoungPaulDanielYoung/internal-audit-skills/skills/foundations/setup-internal-audit-skills/SKILL.md

setup-internal-audit-skills

Skill2 starsChanged 41 days agoMITView on GitHub

Configure the project for internal audit skills by setting up the workspace. Run once before first use of the other internal audit skills.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create CLAUDE.md if it doesn't exist.

    line 19
  2. Very specific:

    If an ## Internal audit skills block already exists in CLAUDE.md, update its contents in-place rather than appending a duplicate.

    line 21
  3. Very specific:

    Create context/ and its subfolder policies/, if they do not exist.

    line 108

≈ “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

Setup Internal Audit Skills

Prepare the workspace with the necessary configuration that the internal audit skills assume exist.

1. Establish the workspace

The current directory is used as the default workspace if no target folder is supplied by the user.

Inspect the existing files relevant to setup before making changes. If existing content conflict with the new setup, resolve the conflict with the user before writing.

2. Create or Update CLAUDE.md

Create CLAUDE.md if it doesn't exist. Otherwise, edit it.

If an ## Internal audit skills block already exists in CLAUDE.md, update its contents in-place rather than appending a duplicate.

## Internal audit skills

How the internal audit skills should consume workspace documentation when performing audit work.

### Before working, read these files in `context/`

- `GLOSSARY.md`
- `METHODOLOGY.md`
- `ORGANIZATION.md`

If any of these files are absent or contain only a blank template, **proceed silently**. Do not flag their absence or blank content alone; don't suggest populating them upfront.

Everything in `context/` is an approved reference maintained by the internal audit department. Agents read it without making edits.

### File structure

The full workspace layout is shown below. Files and directories are created as needed by the relevant skills.

```text
/
├── CLAUDE.md
├── context/
│   ├── GLOSSARY.md
│   ├── METHODOLOGY.md
│   ├── ORGANIZATION.md
│   └── policies/
└── engagements/
    └── <year>/
        └── <name>/
            ├── ENGAGEMENT.md
            ├── document-requests/
            │   ├── REQUESTS.md
            │   └── received/
            ├── fieldwork/
            ├── planning/
            │   ├── risk-assessment.md
            │   ├── risk-and-control-matrix.md
            │   ├── preliminary-survey.md
            │   └── planning-memo.md
            ├── reporting/
            └── walkthroughs/
```

### Resolve material ambiguity with the auditor

When ambiguity about context, assumptions, preferences, or decisions could materially change the work, call the `shared-understanding` skill and resolve it with the auditor before producing the affected part. Continue work the ambiguity doesn't affect.

Clarification through `shared-understanding` settles the current work only; it does not approve changes to `context/METHODOLOGY.md` or `context/ORGANIZATION.md`.

### Use the glossary's vocabulary

When your output names an audit concept, use the term as defined in `context/GLOSSARY.md`.

If a term or concept arises that would be useful to define in the glossary, suggest it to the auditor.

### Follow the documented methodology

When the work involves something addressed by the audit methodology, follow the relevant guidance and requirements in `context/METHODOLOGY.md`. Apply what is documented rather than inventing a different procedure or convention.

If the methodology does not address something needed for the work, don't treat the absence as permission to invent it. Note the genuine methodology gap and resolve uncertainty with the auditor by calling the `shared-understanding` skill.

### Tailor work to the organization

Use the organization's name, description, and industry to tailor your work to the organization.

Use the organization's name when it is natural and useful.

### Consult context documents on demand

When the work depends on a policy, find it at `context/policies/` and read only the documents that bear on the work.

Each context document opens with front matter giving its title, description, and effective date; cite it with those.

### Engagements

Engagements live in `engagements/<year>/<name>/`. Before producing engagement work, resolve the engagement named by the auditor and read its `ENGAGEMENT.md`. Ask if the selection is ambiguous. Keep engagement-specific outputs within that engagement's folder.

### Track document requests

Call the `document-request-list` skill when starting work on an engagement, and whenever work reveals missing material.

3. Create the context folder

Create context/ and its subfolder policies/, if they do not exist. Leave existing folders and their contents unchanged.

4. Create files from templates

Copy each template from this skill's folder to its destination. If the file already exists, skip it and leave it unchanged.

TemplateDestination
GLOSSARY.mdcontext/GLOSSARY.md
METHODOLOGY.mdcontext/METHODOLOGY.md
ORGANIZATION.mdcontext/ORGANIZATION.md

5. Report

Tell the user setup is complete and the internal audit skills will now read from the files in context/.

Explain that context/GLOSSARY.md, context/METHODOLOGY.md, and context/ORGANIZATION.md are blank templates for the user to complete and maintain. The user is responsible for keeping the information in these files accurate and up to date.

Explain that organizational policies converted to Markdown should be added to context/policies/.

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.