incu-way-arch-assessment
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes source code, documentation, and Git diffs from the target repository. This creates a surface where malicious content within the repository could potentially influence the agent's assessment logic.\n
- Ingestion points: Processes repository files, modules,
git diffoutput, and documentation files such asPRD.mdor ADRs.\n - Boundary markers: The skill lacks specific delimiters or instructions to ensure that content read from the codebase is treated strictly as data rather than instructions.\n
- Capability inventory: The skill is capable of writing to the local file system (assessment reports) and performing Git branch/worktree management.\n
- Sanitization: No explicit sanitization or validation of codebase content is performed before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill executes local shell commands to manage the assessment workflow and ensure isolation.\n
- Evidence: Instructions include the use of
git switch,git pull,git fetch, andgit worktree addto create dedicated assessment branches.\n - Context: These commands are used to maintain a clean workspace for findings and are standard for developer-centric agent skills.
Audit Metadata