code-assistant
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes repository files to maintain code consistency, which presents a surface for indirect prompt injection if repository files contain adversarial instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads existing repository files in the current project to match code patterns.
- Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when reading project files.
- Capability inventory: The skill is capable of generating file diffs and suggesting shell commands (building, linting, installing packages) via the
suggest-shelltool. - Sanitization: No specific sanitization or validation of the ingested file content is mentioned.
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to suggest shell commands for environment setup, package management, and build processes (e.g.,
pnpm create next-app,pnpm build,npm install @skillcodex/skills).
Audit Metadata