skills/bjesuiter/skills/jb-bgproc/Gen Agent Trust Hub

jb-bgproc

Warn

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the bgproc utility globally via the Bun package manager using the command 'bun i -g bgproc' as specified in the metadata.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of arbitrary commands in the background via the 'bgproc start' command, which allows for persistent subprocess execution controlled by the agent or user input.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it is designed to ingest and parse output from background processes (logs and status JSON) without clear boundary markers or sanitization. 1. Ingestion points: Output from 'bgproc logs' and 'bgproc status'. 2. Boundary markers: Absent. 3. Capability inventory: Background subprocess execution and management. 4. Sanitization: Absent; instructions suggest direct parsing of JSON output.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 23, 2026, 07:49 PM