python-changelog-fragments

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to read existing changelog fragments to match their tone and structure, creating a surface for indirect prompt injection from untrusted project files.
  • Ingestion points: Files located in changelogs/fragments/ and changelog.d/.
  • Boundary markers: The instructions do not define delimiters or provide warnings to ignore embedded instructions in the ingested files.
  • Capability inventory: The agent has Bash access to execute package managers and changelog tools, alongside file writing capabilities.
  • Sanitization: The skill lacks explicit sanitization or validation steps for the content read from fragment files.
  • [EXTERNAL_DOWNLOADS]: The skill provides procedures for installing and running third-party Python packages including antsibull-changelog, towncrier, and scriv from external registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 09:24 PM