The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from GitHub issues and pull requests, creating a surface for indirect prompt injection attacks.
Ingestion points: Content is pulled from external GitHub repositories using the ghcrawl sync, ghcrawl refresh, and ghcrawl threads commands as described in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters or instructions for the agent to treat the ingested GitHub data as untrusted content.
Capability inventory: The skill has the capability to execute ghcrawl CLI commands, which interact with a local SQLite database and make network requests to GitHub APIs.
Sanitization: There are no instructions for sanitizing or escaping the data retrieved from GitHub before it is processed by the agent.
[COMMAND_EXECUTION]: The skill relies on the execution of the ghcrawl command-line utility for all its operations, including local database inspection, synchronization with remote repositories, and performing embeddings and clustering.

ghcrawl-cluster-operator