changelog-generator

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts (scripts/generate_changelog.py and scripts/commit_linter.py) to process commit data and generate or update the CHANGELOG.md file. These operations are performed via the shell.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from git commit messages and pull request metadata. Malicious content within a commit could attempt to influence the agent's summarization or "rewriting for user audience" logic.
  • Ingestion points: Commit messages and git logs fetched from the repository or provided via stdin/file input (referenced in SKILL.md and README.md).
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to isolate or ignore potentially malicious instructions embedded within the commit history.
  • Capability inventory: The skill possesses the capability to execute shell commands (python3) and perform file system writes (writing to CHANGELOG.md), as described in SKILL.md.
  • Sanitization: No sanitization or validation of the commit message content is specified in the documentation to mitigate instruction injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:07 PM
Security Audit — agent-trust-hub — changelog-generator