Skip to content
Back to skills

Document Agent

ASecurity

Generate a documentation page for a council agent

  • 3 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added May 26, 2026
documentationgodocumentation

Security analysis

A100/100

Scanned May 27, 2026

npx -y skills add golemfoundation/octant-council-builder --skill document-agent --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Document Agent?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Document Agent
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/golemfoundation-document-agent/badge)](https://www.skillsdirectory.com/skills/golemfoundation-document-agent)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
description: Generate a documentation page for a council agent
argument-hint: <agent-name>
allowed-tools:
  - Read
  - Write
  - Glob
model: sonnet
context: fork
user-invocable: false
---

# Generate Agent Documentation

Create a documentation page for a council agent, covering what it does, how it works, and how to customize it.

## Input

`$ARGUMENTS` is the agent name (e.g., `data-audits`, `eval-governance`, `synth-debate`).

## Process

### Step 1: Read inputs

```
Read agents/$ARGUMENTS.md                → agent definition
Read research/$ARGUMENTS.md     → domain context (if exists)
```

Extract from the agent definition:
- Name and description (frontmatter)
- Wave type (from prefix: data/eval/synth)
- Process steps
- Dimensions or sources
- Output format

### Step 2: Write documentation

Write `docs/$ARGUMENTS.md`:

```markdown
# $ARGUMENTS

**Wave:** [1 — Data Gathering | 2 — Evaluation | 3 — Synthesis]
**Role:** [description from frontmatter]

## What This Agent Does

[2-3 sentences explaining the agent's purpose in plain language. What question does it answer? What data does it produce or what judgment does it render?]

## [Data Sources | Scoring Dimensions | Synthesis Method]

[For data agents: table of sources with URLs and what each provides]
[For eval agents: table of dimensions with descriptions and score calibration]
[For synth agents: description of synthesis methodology]

### [Source/Dimension 1]

[1-2 sentences on what this covers and why it matters]

### [Source/Dimension 2]

...

## Output

The agent writes its output to:
- **Data agents:** `council-out/{project}/data/{name}.md`
- **Eval agents:** `council-out/{project}/eval/{name}.md`
- **Synth agents:** `council-out/{project}/REPORT.md`

### Example Output Structure

[Show the key sections of the output format from the agent definition, condensed]

## Customization

### Changing [Sources | Dimensions | Method]

Edit `agents/$ARGUMENTS.md` and modify the [sources table | dimensions table | synthesis section].

### Common Modifications

- [Modification 1: e.g., "Add a new data source by adding a row to the Sources table and a search step in the Process"]
- [Modification 2: e.g., "Change score calibration by editing the calibration table at the bottom"]
- [Modification 3: e.g., "Adjust dimension weights by modifying the composite score calculation"]

## Dependencies

- **Reads from:** [what data this agent needs — e.g., "All Wave 1 data files" or "N/A for data agents"]
- **Consumed by:** [what reads this agent's output — e.g., "All Wave 2 evaluators" or "synth-chair"]
```

### Step 3: Verify

Read back `docs/$ARGUMENTS.md` to confirm:
- All sections are present
- Wave type is correct
- Source/dimension names match the agent definition
- Customization instructions are actionable

## Output

The documentation page at `docs/$ARGUMENTS.md`.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.