The Agent Skills Directory

[SAFE]: No malicious patterns or security threats were detected in the skill's instructions or metadata. The skill performs legitimate operations related to its stated purpose of generating changelogs from git history.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill interacts with the local git repository and project-specific files like CHANGELOG_STYLE.md. It does not perform any network operations to non-whitelisted domains or access sensitive system files such as credentials or private keys.
[PROMPT_INJECTION]: The instructions are focused on guiding the agent through the task of summarization and categorization. No bypass techniques, role-play injections, or instructions to ignore safety filters were found in the skill content.
[INDIRECT_PROMPT_INJECTION]: This skill processes untrusted data from git commit messages, which presents a surface for indirect prompt injection.
Ingestion points: Git commit history and repository-based configuration files (e.g., CHANGELOG_STYLE.md).
Boundary markers: None explicitly defined in the instructions to delimit commit messages.
Capability inventory: Execution of git commands and writing to local files (CHANGELOG.md).
Sanitization: The skill does not explicitly instruct the agent to sanitize or escape commit content before processing.

changelog-generator