knip
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection through its ingestion of codebase data.
- Ingestion points: The agent analyzes repository source files and the text output generated by the Knip CLI (SKILL.md).
- Boundary markers: There are no specified delimiters or guardrail instructions to ensure the agent ignores instructions embedded within the analyzed code or tool output.
- Capability inventory: The skill allows for the execution of shell commands and filesystem modifications when using the
--fixparameter (SKILL.md, references/cli-reference.md). - Sanitization: The skill does not implement sanitization or filtering for the external codebase data it processes.
- [COMMAND_EXECUTION]: The skill provides workflows for executing the
knipcommand-line tool and local Python evaluation scripts (SKILL.md, references/cli-reference.md, evals/evals.json).
Audit Metadata