create-wiki
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill investigates environment variables and configuration files to document the project's setup, which is a standard documentation task. No credentials are exfiltrated.
- [PROMPT_INJECTION]: The skill ingests untrusted codebase data (source code, manifests, and DESIGN.md) to populate the wiki template. This creates a surface for indirect prompt injection. 1. Ingestion points: Reads all repository files as directed by the investigation step in SKILL.md. 2. Boundary markers: Absent; no specific instructions are provided to ignore embedded instructions in the codebase. 3. Capability inventory: File read (entire repository) and File write (.docs/index.html) as detailed in SKILL.md. 4. Sanitization: Absent; the skill does not instruct the agent to sanitize or escape content before it is interpolated into references/template.html.
Audit Metadata