Agent skill · Audited from source

core-fix-skill-docs

skill · actionbook/rust-skillsLicence not declaredAudited 24 Oct 2026 · source f4ff8c82542d

★ 1,541pushed 62 days ago
✓SAFEGrade B · 89/100 · 24 Oct 2026 · f4ff8c8
1,541repo stars · whole actionbook/rust-skills

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit f4ff8c82542d and found no filesystem, network or shell access and no hard-coded secrets. What it does: Internal maintenance support for checking and fixing generated Rust skill documentation references. Use only when explicitly invoked by /fix-skill-docs.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “core-fix-skill-docs (actionbook/rust-skills)”. Retrieved 24 Oct 2026. https://modelsatlas.com/skills/actionbook/core-fix-skill-docs

Trust audit

Audit 0.4.1 · source read 24 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium1 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 13,985 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS1
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass, no license.

Findings (1)

LOWInventory / provenanceinv.no_license
F-13 Inventory / provenanceCWE-1104Layer L0
Why
no LICENSE file and no repo licence
Fix
add a licence

Audit 0.4.1 · source sha f4ff8c82542d · 24 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisheractionbook GitHub organizationOwner of the repository this skill is read from
  • Source folderactionbook/rust-skills/skills/core-fix-skill-docsThe exact folder we audited
  • LicenceNot declaredAs the repository states it
  • Commit pinned for this verdictf4ff8c82542dA new audit runs when the source changes
  • Repository statusNot archivedLast push 62 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Internal maintenance support for checking and fixing generated Rust skill documentation references. Use only when explicitly invoked by /fix-skill-docs.

---
name: core-fix-skill-docs
description: "Internal maintenance support for checking and fixing generated Rust skill documentation references. Use only when explicitly invoked by /fix-skill-docs."
disable-model-invocation: true
argument-hint: "[crate_name] [--check-only]"
context: fork
agent: general-purpose
---

# Fix Skill Documentation

> **Version:** 2.1.0 | **Last Updated:** 2025-01-27

Check and fix missing reference files in dynamic skills.

## Usage

```
/fix-skill-docs [crate_name] [--check-only] [--remove-invalid]
```

**Arguments:**
- `crate_name`: Specific crate to check (optional, defaults to all)
- `--check-only`: Only report issues, don't fix
- `--remove-invalid`: Remove invalid references instead of creating files

## Execution Mode Detection

**CRITICAL: Check if agent infrastructure is available.**

This skill can run in two modes:
- **Agent Mode**: Uses background agents for documentation fetching
- **Inline Mode**: Executes directly using agent-browser CLI or WebFetch

---

## Agent Mode (Plugin Install)

**When agent infrastructure is available, use background agents for fetching:**

### Instructions

#### 1. Scan Skills Directory

```bash
# If crate_name provided
skill_dir=~/.claude/skills/{crate_name}

# Otherwise scan all
for dir in ~/.claude/skills/*/; do
    # Process each skill
done
```

#### 2. Parse SKILL.md for References

Extract referenced files from Documentation section:

```markdown
## Documentation
- `./references/file1.md` - Description
```

#### 3. Check File Existence

```bash
if [ ! -f "{skill_dir}/references/{filename}" ]; then
    echo "MISSING: {filename}"
fi
```

#### 4. Report Status

```
=== {crate_name} ===
SKILL.md: OK
references/:
  - sync.md: OK
  - runtime.md: MISSING

Action needed: 1 file missing
```

#### 5. Fix Missing Files (Agent Mode)

Launch background agent to fetch documentation:

```
Task(
  subagent_type: "general-purpose",
  run_in_background: true,
  prompt: "Fetch documentation for {crate_name}/{module} from docs.rs.
           Use agent-browser CLI to navigate to https://docs.rs/{crate_name}/latest/{crate_name}/{module}/
           Extract the main documentation and save to ~/.claude/skills/{crate_name}/references/{module}.md"
)
```

---

## Inline Mode (Skills-only Install)

**When agent infrastructure is NOT available, execute directly:**

### Step 1: Scan Skills Directory

```bash
# List all skills
ls ~/.claude/skills/

# Or check specific skill
ls ~/.claude/skills/{crate_name}/
```

### Step 2: Parse SKILL.md for References

Read SKILL.md and extract all `./references/*.md` patterns:

```bash
# Using Read tool
Read("~/.claude/skills/{crate_name}/SKILL.md")

# Look for lines like:
# - `./references/sync.md` - Sync primitives
# - `./references/runtime.md` - Runtime configuration
```

### Step 3: Check File Existence

```bash
# Check each referenced file
for ref in references; do
  if [ ! -f "~/.claude/skills/{crate_name}/references/${ref}.md" ]; then
    echo "MISSING: ${ref}.md"
  fi
done
```

### Step 4: Report Status

Output format:
```
=== {crate_name} ===
SKILL.md: OK
references/:
  - sync.md: OK
  - runtime.md: MISSING

Action needed: 1 file missing
```

### Step 5: Fix Missing Files (Inline)

For each missing file:

**Using agent-browser CLI:**
```bash
agent-browser open "https://docs.rs/{crate_name}/latest/{crate_name}/{module}/"
agent-browser get text ".docblock"
# Save output to ~/.claude/skills/{crate_name}/references/{module}.md
agent-browser close
```

**Using WebFetch fallback:**
```
WebFetch("https://docs.rs/{crate_name}/latest/{crate_name}/{module}/",
         "Extract the main documentation content for this module")
```

Then write the content:
```bash
Write("~/.claude/skills/{crate_name}/references/{module}.md", <fetched_content>)
```

### Step 6: Update SKILL.md (if --remove-invalid)

If `--remove-invalid` flag is set and file cannot be fetched:

```bash
# Read current SKILL.md
Read("~/.claude/skills/{crate_name}/SKILL.md")

# Remove the invalid reference line
Edit("~/.claude/skills/{crate_name}/SKILL.md",
     old_string="- `./references/{invalid_file}.md` - Description",
     new_string="")
```

---

## Tool Priority

1. **agent-browser CLI** - Primary tool for fetching documentation
2. **WebFetch** - Fallback if agent-browser unavailable
3. **Edit SKILL.md** - For removing invalid references (--remove-invalid only)

---

## Examples

### Check All Skills (--check-only)

```bash
/fix-skill-docs --check-only

# Output:
=== tokio ===
SKILL.md: OK
references/:
  - sync.md: OK
  - runtime.md: MISSING
  - task.md: OK

=== serde ===
SKILL.md: OK
references/:
  - derive.md: OK

Summary: 1 file missing in 1 skill
```

### Fix Specific Crate

```bash
/fix-skill-docs tokio

# Fetches missing runtime.md from docs.rs
# Reports success
```

### Remove Invalid References

```bash
/fix-skill-docs tokio --remove-invalid

# If runtime.md cannot be fetched:
# Removes reference from SKILL.md instead
```

---

## Error Handling

| Error | Cause | Solution |
|-------|-------|----------|
| Agent not available | Skills-only install | Use inline mode |
| Skills directory empty | No skills installed | Run /sync-crate-skills first |
| docs.rs unavailable | Network issue | Retry or use --remove-invalid |
| Permission denied | Directory issue | Check ~/.claude/skills/ permissions |
| Invalid SKILL.md format | Corrupted skill | Re-generate skill |

Excerpt, first 5,409 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit f4ff8c8

Download folder ↗
git clone https://github.com/actionbook/rust-skills
git -C rust-skills checkout f4ff8c82542d

The author documents no install command, so this uses the repository directly. Our verdict covers commit f4ff8c82542d only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source f4ff8c82542d

README badge

For maintainers · shows the latest audit

ModelsAtlas: SAFE BMODELSATLASSAFE · B
[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/actionbook/core-fix-skill-docs/badge.svg)](https://modelsatlas.com/skills/actionbook/core-fix-skill-docs/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Add this badge to your README

[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/actionbook/core-fix-skill-docs/badge.svg)](https://modelsatlas.com/skills/actionbook/core-fix-skill-docs/)

Questions about core-fix-skill-docs

What does the core-fix-skill-docs skill do?
Rust Developer AI Assistance System — Meta-Problem-Driven Knowledge Indexing
Is core-fix-skill-docs safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can core-fix-skill-docs access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (f4ff8c82542d), read on 2026-10-24. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.