The Agent Skills Directory

[COMMAND_EXECUTION] (LOW): The tool utilizes child_process.spawn and child_process.exec within lib/doc-generator.js and static-assets/server.js to manage a local documentation server and automate opening the system's web browser. Evidence: spawn('node', [serverScript], ...) and exec("${startCmd} ${url}"). These actions are necessary for the skill's primary function and are constrained to the local environment.\n- [PROMPT_INJECTION] (LOW): The skill possesses a surface for Indirect Prompt Injection (Category 8) because its workflow depends on ingesting and summarizing untrusted data from project repositories.\n
Ingestion points: Local codebase files are analyzed using shell utilities like ls, rg, sed, and git.\n
Boundary markers: Absent; the skill does not wrap ingested external content in security delimiters to distinguish it from system instructions.\n
Capability inventory: The skill has filesystem write access (fs.writeFileSync in lib/doc-generator.js) and can execute local Node.js processes and shell commands.\n
Sanitization: The provided JavaScript tools do not perform automated HTML sanitization or XSS filtering, relying instead on instructions within SKILL.md for the AI agent to escape characters manually.

skill-system-analyst