The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, which presents an attack surface for indirect prompt injection.
Ingestion points: Fetches pull request bodies, review findings, and issue content using the github-facts.mjs helper and fetchSelectedReview in adapters.ts.
Boundary markers: Utilizes structured delimiters such as  and  to separate machine-readable state from prose.
Capability inventory: Capable of repository mutation (squash-merging, labeling, closing tickets) via the gh CLI and performing file writes to the local checkout.
Sanitization: Implements a strict validation protocol where verification claims are cross-checked against a local configuredCommands allowlist, and requires execution receipts to authenticate passing status.
[COMMAND_EXECUTION]: The skill uses the gh (GitHub) CLI tool to perform repository management operations.
Evidence: The gh-fix-transport.ts and github-facts.mjs scripts use spawnSync and execFileSync to execute gh api calls for merging, commenting, and updating issue states.
[EXTERNAL_DOWNLOADS]: The skill references and requires the installation of additional tools from the author's verified GitHub repository.
Evidence: INSTALL.md and SKILL.md provide instructions to install the fix-this and unslopify skills from the RuralNative/RuralNative-SKILLS repository.

fix-this