job-search-copilot
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill's functionality for processing external data presents a surface for indirect prompt injection.
- Ingestion points: The skill ingests untrusted data from LinkedIn job listings (via scraping) and Gmail alerts (via
/job emailas described inSKILL.md). - Boundary markers: No specific boundary markers or instructions to ignore embedded instructions within processed data are present in the skill files.
- Capability inventory: The agent has the capability to write files (materials generation), execute shell scripts (
~/agent-tools/), and perform network operations (viajob-scraper.js). - Sanitization: There is no evidence of sanitization or filtering of the content retrieved from external sources before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill relies on local script execution for its automation pipeline.
- Evidence: Multiple commands in
SKILL.md(e.g.,/job qualify,/job scrape) execute Node.js scripts located in~/agent-tools/and~/.claude/skills/. - Behavior: The
job-qualify.jsscript performs local file system operations, reading from~/.claude/skills/job-search-copilot/configand writing to~/.claude/state/job-listings-cache.json.
Audit Metadata