quality-auditor
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill evaluates external tools and codebases, creating a vulnerability to indirect prompt injection if the audited content contains malicious instructions. \n
- Ingestion points: The agent is instructed to 'Read all documentation', 'Examine the codebase', and parse specific registry files like 'META/registry.json' as defined in 'SKILL.md'.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded prompts (e.g., in markdown comments or code strings) are provided to the agent for the analysis phase.\n
- Capability inventory: The agent has the ability to execute shell commands ('bash' tools) and modify files (as per 'manifest.yaml'), which could be exploited if an injected prompt successfully overrides the agent's behavior.\n
- Sanitization: The skill lacks any mechanism to sanitize or validate the content of the files being audited before they are processed by the agent's context.\n- [COMMAND_EXECUTION]: In 'SKILL.md', the 'Phase 0: Resource Completeness Check' explicitly directs the agent to execute shell commands including 'npm run', 'ls', 'grep', 'wc', and 'jq'. While these are used for project integrity verification, they constitute an active command execution surface that relies on the state of the local environment and the contents of the project being audited (e.g., the 'package.json' script for 'npm run').
Audit Metadata