comment-slayer
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from external files, commits, and pull requests, which are potentially untrusted data sources. An attacker could embed malicious instructions within code comments aimed at manipulating the agent's behavior during the audit process.
- Ingestion points: Processes file systemic content, git diffs, and pull request data as specified in the Scope section of SKILL.md.
- Boundary markers: The skill does not define structural delimiters or provide instructions to the agent to treat the content of comments as data rather than instructions.
- Capability inventory: The agent has permissions to read the filesystem and can perform file-write operations (deleting or replacing text) when the fix argument is provided.
- Sanitization: No explicit sanitization or filtering of the analyzed prose is performed before the agent processes the text to determine the verdict.
Audit Metadata