test-effort-estimation
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from git history and repository files to classify change shapes and estimate effort. 1. Ingestion points: Git log output and source file contents are processed via the instructions in references/change-shape-classifier.md and scripts/classify-change-shape.py. 2. Boundary markers: The skill lacks delimiters or specific instructions to the agent to ignore potentially malicious instructions embedded in the analyzed code or commit messages. 3. Capability inventory: The skill requests execution of shell commands (git log, git diff) and the creation and execution of a Python script to analyze the repository data. 4. Sanitization: No sanitization is applied to the data before it is parsed and processed by the agent.
Audit Metadata