The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local bash script (scripts/docs_qc.sh) to perform structural linting and quality checks on documentation files. The script uses standard system utilities such as find, ripgrep (rg), and wc to verify frontmatter, identify potential parsing issues in Markdown tables, and check for appropriate headings in diagrams. The script performs read-only operations and does not access external networks.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process content from the repository that may be controlled by third parties (e.g., documentation files or source code from pull requests).
Ingestion points: The skill reads multiple files, including docs/*.mdx, src/types.ts, schema/isol8.config.schema.json, and general source files under src/ to gather context.
Boundary markers: There are no explicit delimiters or system instructions defined in the workflow to prevent the agent from following malicious instructions embedded within the processed documentation or code files.
Capability inventory: The agent has the capability to execute the local docs_qc.sh script and is expected to modify or create documentation files based on the ingested content.
Sanitization: The skill does not implement any validation, escaping, or filtering of the ingested content before it is processed by the AI agent.

docs-quality