The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell operations using gh (GitHub CLI) and git to manage documentation updates. This includes fetching repository content, writing to temporary files, branching, committing, and opening pull requests in ElevenLabs repositories.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from a GitHub repository and the vendor's website to generate documentation edits.
Ingestion points: Reads changelog files from the elevenlabs-dx repository and documentation from elevenlabs.io.
Boundary markers: No explicit delimiters are specified for isolating ingested content, although a step-by-step verification process is required.
Capability inventory: The skill can write to the filesystem, execute Git commands, and interact with the GitHub API to modify repositories and create PRs.
Sanitization: Content is checked for accuracy against documentation, but there is no programmatic sanitization of the input data before it influences agent behavior.
[DYNAMIC_EXECUTION]: The skill provides a shell command template for gh api that interpolates the CHANGELOG_DATE variable directly from user input or automation triggers. This pattern presents a command injection risk if the variable is not strictly validated or escaped before execution.

update-skills-from-changelog