hunter-scout

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill connects to multiple well-known job board platforms (Greenhouse, Lever, Ashby, Workday, and Eightfold) to fetch live job listings. These network operations are core to the skill's documented functionality and target official public APIs.
  • [DATA_EXPOSURE]: The skill reads the user's profile (hunter/profile.json) to perform local scoring of job listings. When performing searches on Workday or Eightfold boards, it sends the user's target job titles as search query parameters. This is a standard and necessary behavior for the intended purpose of the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests job descriptions from external sources which could theoretically contain malicious instructions. To mitigate this risk, the skill implements a sanitization function (html_text) in jobs.py that strips HTML tags, scripts, and style blocks before the content is processed or displayed to the agent.
  • Ingestion points: Job descriptions are fetched from external board APIs in jobs.py via the Fetcher.get method.
  • Boundary markers: Job listings are presented in a structured list format, though they do not use specific security-oriented delimiters for the description text.
  • Capability inventory: The skill has the capability to write to local files (hunter/jobs.json, hunter/pipeline.json) and perform HTTP GET/POST requests.
  • Sanitization: The html_text utility removes executable HTML elements and scripts from the ingested text.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 08:22 PM