The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Bash to execute git commands (e.g., git log, git describe) for extracting commit history. It also performs project structure detection by checking for the existence of specific build files (e.g., pom.xml, package.json, build.gradle.kts) and reading repository metadata. These actions are necessary for the skill's stated purpose and are performed locally.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of external commit message data during the summarization process.
Ingestion points: Commit metadata and messages are ingested from the repository's history via git log commands as detailed in reference/changelog-workflow.md.
Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore potentially malicious instructions embedded within the commit messages being summarized.
Capability inventory: The skill utilizes Bash, Read, Write, and Edit tools, which could be leveraged if an injection is successful.
Sanitization: No sanitization or filtering logic is provided for the commit strings before they are presented to the agent for rewriting into user-facing language.

changelog-generator