no-comments
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon untrusted data from external sources, which can be exploited to bypass instructions.
- Ingestion points: Processes caller-provided files and diffs, as well as the output reports from the
comment-sickosubagent (SKILL.md). - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings when processing the contents of files or subagent reports.
- Capability inventory: The agent has the capability to spawn subagents and perform file modifications, including deletions and code "fixes" based on the ingested content.
- Sanitization: The skill relies on natural language reasoning (e.g., "Reject application-code edits") to filter inputs rather than programmatic sanitization or strict schema validation.
Audit Metadata