lint
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses dynamic context injection to run 'pnpm lint:eslint 2>&1' at load time. This is a legitimate use of project-specific tooling to populate the agent's context with immediate linting results.
- [PROMPT_INJECTION]: There is a potential for indirect prompt injection. 1. Ingestion points: Untrusted data enters the agent context via the output of the linter command in SKILL.md. 2. Boundary markers: The output is enclosed in <eslint_output> tags, which helps the agent identify the data but does not fully prevent instruction override. 3. Capability inventory: The agent has the power to read and modify files (Edit) and run additional bash commands. 4. Sanitization: No sanitization is performed on the ingested linter output, which could potentially contain attacker-controlled strings (e.g., in file names or crafted lint messages) designed to influence the agent's behavior during the fix process.
Audit Metadata