skills/ewurch/bragger/applications/Gen Agent Trust Hub

applications

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on executing a CLI tool named bragger to perform data operations. User-supplied input from various fields, such as company names, roles, and notes, is passed directly as arguments to this command-line interface.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted content from external job descriptions and files.
  • Ingestion points: Untrusted data enters the agent context via the jd_content field, the --jd-file path, and the applications.jsonl storage file.
  • Boundary markers: The instructions lack defined delimiters or specific 'ignore embedded instructions' warnings to prevent the agent from obeying commands hidden within job description text.
  • Capability inventory: The skill can execute shell commands (bragger) and perform file system read/write operations on the application tracker.
  • Sanitization: No sanitization, escaping, or validation of the ingested job description content or file paths is described.
  • [EXTERNAL_DOWNLOADS]: The skill handles external URLs for job descriptions and company websites, and it includes logic to fetch and extract content from these remote sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 10:34 PM