no-comments
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted source code and git diffs provided by the user.
- Ingestion points: The skill accesses the 'caller's files or diff' as specified in the Scope section to perform its primary function.
- Boundary markers: There are no instructions provided to the agent to treat the code content as strictly data or to ignore embedded instructions (e.g., in comments) that might attempt to override the skill's logic.
- Capability inventory: The skill possesses significant capabilities including spawning subagents, performing file modifications (deleting code, implementing fixes), and executing complex tools like
/architectand/why. - Sanitization: No explicit sanitization or validation of the input source code is performed prior to the agent processing it for comment removal or refactoring.
- [COMMAND_EXECUTION]: The skill utilizes several built-in agent commands and subagent orchestration features.
- Evidence: The instructions explicitly direct the agent to use the
/whycommand to trace symbols and the/architectcommand to sketch code fixes. It also orchestrates a 'Task' with a specific 'subagent_type' (comment-sicko).
Audit Metadata