maple-audit
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and act upon untrusted source code from the repository being audited. A malicious codebase could potentially contain comments or patterns designed to influence the agent's audit findings or fix logic.
- Ingestion points: The skill reads application source code, workspace manifests, and dependency files across the repository (SKILL.md, Step 1 & 2).
- Boundary markers: There are no specific delimiters or instructional guardrails provided to the agent to treat the code content as untrusted data separate from its instructions.
- Capability inventory: The skill has the capability to write modifications to the local filesystem (SKILL.md, Step 5) and execute shell commands (SKILL.md, Step 6).
- Sanitization: No validation, escaping, or sanitization of the code content is performed before it is analyzed or used to generate fixes.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the project's own development and build commands during the verification phase (SKILL.md, Step 6). While this is a functional requirement for verifying instrumentation fixes, it involves executing arbitrary scripts defined within the untrusted project environment.
Audit Metadata