The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill possesses a surface area for indirect prompt injection by design, as its primary function is to ingest and analyze documentation from external repositories.
Ingestion points: Files matching patterns AGENTS.md, CONTRIBUTING.md, README*, docs/**, **/*.md, **/*.mdx, **/*.mdc, **/*.rst, and **/*.rsc are read by the inventory-agent, docs-framework-agent, and governance-agent sub-agents.
Boundary markers: Absent. The instructions do not mandate the use of delimiters or 'ignore' instructions when interpolating repository content into agent prompts.
Capability inventory: The skill uses Read, Glob, Grep, and LS tools. It also supports an apply-fixes mode which implies file-writing capabilities to the local repository based on the analysis of ingested documentation.
Sanitization: Absent. There are no instructions to escape or validate the content of the documentation before it is processed by the LLM sub-agents.

technical-documentation