The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of untrusted external data:
Ingestion points: The agent reads git commit history and merged Pull Request data to generate release notes and changelog entries (SKILL.md).
Boundary markers: None; the skill does not specify delimiters or instructions to ignore potential commands embedded in commit messages or PR descriptions.
Capability inventory: The skill can perform sensitive operations including git push, npm publish, and executing the local script ./scripts/deploy-docs.sh (SKILL.md).
Sanitization: There is no mention of sanitizing or validating the content retrieved from git history before it is interpolated into prompts or used in the release drafting process.
[COMMAND_EXECUTION]: The skill relies on shell command execution to perform its primary functions:
Executes npm run registry:build to perform build tasks.
Uses git commit -am and git push to modify and upload repository content.
Executes a local shell script ./scripts/deploy-docs.sh for documentation deployment.
Runs npm publish to distribute the package to the public registry.

release