web-crawl
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest external web content, which presents a surface for indirect prompt injection attacks.
- Ingestion points: Untrusted external data enters the context via URLs fetched by
harness/scripts/web-crawl.py(referenced inSKILL.md). - Boundary markers: The skill does not define specific delimiters or warnings for the agent to ignore instructions embedded in the crawled markdown.
- Capability inventory: The skill utilizes subprocess calls to
python3for script execution and writes files to theraw/directory (defined inSKILL.md). - Sanitization: The skill performs HTML-to-markdown conversion, but does not implement sanitization for natural language instructions embedded in the text.
- [COMMAND_EXECUTION]: The skill executes a local Python script (
harness/scripts/web-crawl.py) using the system's python3 interpreter to perform fetching and markdown conversion. - [EXTERNAL_DOWNLOADS]: The skill performs outbound network requests to fetch content from user-specified URLs. This is a core functional requirement but involves interaction with external, untrusted network endpoints.
Audit Metadata