Skills
skills/mikecfisher/agent-debug-mode/debug-fixed/Gen Agent Trust Hub

debug-fixed

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes bun ./scripts/stop-collector.mjs to stop a local background process and suggests rm -rf .debug for directory cleanup.
  • [COMMAND_EXECUTION]: The script scripts/stop-collector.mjs uses process.kill() to send SIGTERM and SIGKILL signals to a process ID retrieved from a local file (.debug/collector.pid).
  • [PROMPT_INJECTION]: The skill uses the $ARGUMENTS placeholder to ingest user notes into the prompt context without the use of boundary markers or specific sanitization, creating a surface for indirect prompt injection if user-provided content contains instructions intended to override the agent's workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 01:09 AM
Security Audit — agent-trust-hub — debug-fixed