agent-safety-preflight
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform git operations and execute a local Python scanner script (agent_preflight_lite.py). These operations are scoped to specific command patterns (git:* and python3:*) as defined in the frontmatter.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it analyzes untrusted content from local repositories to determine safety risks.
- Ingestion points: Local repository files, including shell scripts, configuration files (.claude/, .github/, .env), and package manifests.
- Boundary markers: The instructions do not define specific delimiters or instructions for the agent to disregard malicious prompts embedded within the scanned files.
- Capability inventory: The skill has access to the Read tool for file inspection and the Bash tool for command execution.
- Sanitization: No specific sanitization or filtering is applied to the repository content before it is processed by the agent.
Audit Metadata