internalized-wiki
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its core functionality of fetching and processing external source material.\n
- Ingestion points: External URLs and source files retrieved via web tools or provided by the user (Ingest workflow).\n
- Boundary markers: The skill uses structured templates for output but lacks explicit delimiters to isolate the agent from instructions potentially embedded in the fetched raw source material.\n
- Capability inventory: The agent has the ability to read and write files within the project root and use web navigation tools.\n
- Sanitization: Instructions focus on cleaning formatting noise and artifacts but do not include measures to detect or neutralize malicious instructions in the source text.\n- [COMMAND_EXECUTION]: The skill performs file system management within the project root (specifically 'raw/' and 'wiki/'). It creates and updates markdown files, directory structures, and append-only logs to maintain the wiki's integrity. These operations are essential for the tool's stated purpose.\n- [EXTERNAL_DOWNLOADS]: The skill is designed to fetch content from the internet using available web tools during the ingestion phase. This behavior is transparent and initiated by the user to populate the knowledge base.
Audit Metadata