fix-mypy

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches information from github.com/Azure and mypy.readthedocs.io. These are well-known and trusted sources for developer documentation and issue tracking.
  • [COMMAND_EXECUTION]: The skill performs several legitimate command-line operations required for code maintenance:
  • Invokes pip install to set up development environments and install package dependencies.
  • Uses git to track changes, identify target files, and create pull requests.
  • Executes azpysdk, a vendor-specific tool, to run MyPy checks in an isolated environment.
  • Uses grep to analyze local source code for existing type patterns.
  • [PROMPT_INJECTION]: The skill possesses an indirect injection surface because it reads and processes data from GitHub issues and local source code.
  • Ingestion points: Reads GitHub issue descriptions and source code files via grep.
  • Boundary markers: None identified.
  • Capability inventory: Includes file writing, package installation via pip, and git operations.
  • Sanitization: Relies on specific instructions to the agent to apply fixes only with '100% confidence' and strictly forbids changing code logic or adding non-typing dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 09:48 AM
Security Audit — agent-trust-hub — fix-mypy