The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to instructions embedded in external data it processes.
Ingestion points: The skill reads git commit logs and code diffs via git log and git diff in the repository (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential prompt-like content found within commit messages.
Capability inventory: The skill has the capability to write and update the CHANGELOG.md file in the repository root.
Sanitization: No sanitization or filtering logic is defined to handle malicious or deceptive text within the git history data.
[COMMAND_EXECUTION]: The skill uses shell commands to interact with the local git repository.
Evidence: Instructions include the execution of git tag, git diff, and git log --oneline to analyze project history (SKILL.md). These are standard operations for the skill's stated purpose.

semver-changelog