The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests and processes issue descriptions from external sources which are not controlled by the skill itself.
Ingestion points: Accesses project issues and milestones via external adapters, and reads docs/issue-tracker.md for guidelines.
Boundary markers: The instructions do not specify the use of delimiters or specific prompts to ignore embedded instructions within issue content.
Capability inventory: The skill has the capability to write and publish release notes back to a tracker when authorized.
Sanitization: There are no explicit instructions for sanitizing or escaping the content of issue descriptions before they are processed or rewritten.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads local documentation and fetches issue metadata. These actions are within the scope of its primary purpose for changelog generation and do not involve hardcoded credentials or unauthorized data transfers.

write-changelog