The Agent Skills Directory

[SAFE]: No malicious patterns, data exfiltration, or unauthorized code execution found. The skill follows established developer workflows for generating documentation.
[PROMPT_INJECTION]: The instructions do not contain attempts to bypass safety filters, override core agent behavior, or extract system prompts.
[DATA_EXFILTRATION]: Access is limited to project-specific metadata (package.json, changesets) and git history required for generating release notes. There are no patterns suggesting the collection or external transmission of sensitive credentials or personal data.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from git logs and changeset files, which represents a potential indirect prompt injection surface. However, because the skill lacks dangerous capabilities such as file-system writes, network operations, or dynamic code execution, this surface is considered low risk and inherent to the skill's primary function.
Ingestion points: .changeset/*.md files, git log, and git diff outputs.
Boundary markers: Absent; the instructions do not specify delimiters to isolate external content from the prompt.
Capability inventory: No subprocess calls, network requests, or file-writing operations are requested or used.
Sanitization: No explicit sanitization or filtering of commit messages or changeset content is defined.

release-summary