skills/skill-manifest-generator/SKILL.md
Generate MANIFEST.json files for Agent Skills, providing content integrity verification, file inventory, and external reference tracking.
npx skillsauth add skillscatalog/registry skill-manifest-generatorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill to generate a MANIFEST.json file for any Agent Skill directory. The manifest provides:
Run the manifest generator on a skill directory:
Generate a manifest for the skill at /path/to/my-skill
Or to verify an existing manifest:
Verify the manifest for /path/to/my-skill
The generator creates a MANIFEST.json file in the skill root containing:
{
"$schema": "https://agentskills.io/schemas/manifest.v1.json",
"manifestVersion": "1.0",
"generatedAt": "2025-01-15T10:30:00Z",
"generator": "skill-manifest-generator/1.0.0",
"skill": {
"name": "my-skill",
"version": "1.0.0"
},
"integrity": {
"algorithm": "sha256",
"hash": "a1b2c3d4..."
},
"files": [...],
"externalReferences": [...],
"structure": {...}
}
Generate a manifest:
User: Generate a manifest for the skill at ./pdf-tools
Agent: I'll generate a MANIFEST.json for the pdf-tools skill...
[Runs generate_manifest.py]
Created MANIFEST.json with:
- 12 files inventoried
- Integrity hash: sha256:abc123...
- 2 external URLs detected
- Max folder depth: 2
Verify a manifest:
User: Verify the manifest for ./pdf-tools
Agent: I'll verify the MANIFEST.json matches the current files...
[Runs generate_manifest.py --verify]
✓ All 12 files match their recorded hashes
✓ No new untracked files found
✓ Integrity hash verified
tools
A brief description of what this skill does
development
Validate Agent Skills against the specification. Checks SKILL.md format, frontmatter fields, naming conventions, and directory structure.
development
Search the Agent Skills Catalog to find skills by keyword, vendor, or category.
development
Run local safety scans on Agent Skills before publishing. Detects secrets, dangerous code patterns, and analyzes required permissions.