boss-resume-downloader

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a Python script (scripts/sync_boss_resumes.py) to execute the boss CLI tool via subprocess.run. This is the core mechanism for interacting with the BOSS platform and is implemented using safe argument list patterns to prevent shell injection vulnerabilities.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it downloads and stores external candidate resumes which the agent is then encouraged to read and analyze for recruitment purposes.
  • Ingestion points: The scripts/sync_boss_resumes.py script writes raw candidate data from the BOSS platform into local resume.md files.
  • Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are added to the generated markdown content to isolate it from the agent's instructions.
  • Capability inventory: The agent has access to the boss CLI (via the wrapper script) and the local filesystem for reading and writing resume data.
  • Sanitization: The script performs basic newline normalization during markdown conversion but does not validate or sanitize the textual content of resumes for potential malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 10:29 AM
Security Audit — agent-trust-hub — boss-resume-downloader