The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a local Node.js script (scripts/render-requirements-map.mjs) to process requirement manifests. The script is invoked by the agent to generate HTML reports from structured JSON data.
[EXTERNAL_DOWNLOADS]: The generated HTML reports include a reference to https://cdn.jsdelivr.net/npm/@tailwindcss/browser@4. This is used to compile CSS in the browser at runtime. The skill documentation explicitly warns users about this external connection and advises against using the tool for highly confidential requirements.
[SAFE]: The skill demonstrates high security standards through several mechanisms:
Strict Validation: The renderer script implements a validate function that strictly checks the JSON manifest against a schema, including ID patterns, node types, and hierarchy rules.
XSS Mitigation: User-provided requirement text is properly escaped using escapeHtml and safeJson utilities before being embedded in the HTML report.
Secure File Handling: The script uses atomic file operations (writing to a staging directory with restricted permissions 0o700 then renaming) and unique directory names via randomUUID() to prevent filesystem conflicts or predictable path attacks.
Transparent Safety: The skill includes a test suite (scripts/tests/run-tests.sh) and dedicated fixtures (unsafe.json) to verify that malicious input is properly escaped and rejected.

requirements-map