The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands including gh, git, and base64 to automate the fetch, commit, and pull request workflow. These are standard tools for repository management and documentation automation.
[EXTERNAL_DOWNLOADS]: The skill fetches data from the vendor's own repository at github.com/elevenlabs/elevenlabs-dx. This is a necessary operation for the skill to identify recent changes and is performed using official API endpoints.
[INDIRECT_PROMPT_INJECTION]: The skill ingests content from an external source (the changelog) and uses it to generate file edits and pull request descriptions, creating a surface for indirect influence.
Ingestion points: Markdown content is fetched from the elevenlabs-dx repository via gh api in SKILL.md (Step 1).
Boundary markers: No specific delimiters or "ignore instructions" tags are used for the ingested changelog text.
Capability inventory: The skill possesses capabilities to write files to /tmp, commit changes to a local branch, push to a remote repository, and open pull requests.
Sanitization: The skill does not explicitly describe sanitization or validation logic for the text extracted from the changelog before it is processed by the agent.

update-skills-from-changelog