The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a bundled Node.js script (scripts/validate-llm-wiki.mjs) to verify the integrity and formatting of the generated documentation. This script is a vendor-provided resource from 'asta-nguyen' and performs static analysis on Markdown files. Technical review confirms the script uses only standard built-in modules (node:fs, node:path) for file system operations and contains no network access or dangerous execution logic.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection because its primary function requires reading and processing untrusted external data from the repository's source code and existing documentation.
Ingestion points: The agent reads source code, tests, repository manifests (AGENTS.md), and existing wiki files in docs/llm/.
Boundary markers: While the instructions provide strict rules for evidence verification, they do not implement explicit delimiters or instructions to ignore potential commands embedded within the data being analyzed.
Capability inventory: The agent has the capability to write files to the documentation directory and execute a local validation script.
Sanitization: There is no evidence of content sanitization or filtering before the external source code or documentation is ingested into the agent's context.

document-wiki