The Agent Skills Directory

[SAFE]: Technical review of the provided Python scripts (changelog_generator.py, version_bumper.py, and release_planner.py) confirms they utilize only the Python standard library for data processing. No evidence of malicious network communication, credential harvesting, or unauthorized file system access was found.- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data (git logs and JSON files) to generate reports. While this constitutes an attack surface for indirect prompt injection, the scripts are limited to text parsing and output generation with no execution capabilities (e.g., no subprocess or eval usage on ingested data), effectively mitigating potential risks.- [EXTERNAL_DOWNLOADS]: Examples provided in the documentation demonstrate integration with official APIs (e.g., GitHub API) using the requests library. These are well-known, trusted services and represent standard developer workflows.

release-manager