The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute local shell and Node.js scripts to automate web browser interactions and data scraping.
Evidence: Executes ../../adapters/boss/launch-chrome.sh, node diagnose-cdp.mjs, node read-boss.mjs, and node read-boss-detail.mjs.
[DATA_EXFILTRATION]: The skill reads and writes to .skill-state.json, which contains sensitive user profile information and job search history.
Evidence: Accessed to retrieve profile.start_tier and update candidate_roles[].
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted job descriptions from external websites while maintaining significant system capabilities.
Ingestion points: External job description (JD) text retrieved via the read-boss-detail.mjs adapter script (SKILL.md).
Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore embedded commands within the JD text.
Capability inventory: The skill has access to full shell execution (Bash(*)) and filesystem write access (Write) (SKILL.md).
Sanitization: Absent; there is no evidence of filtering or escaping external content before the agent analyzes the data.

skill-scan