cleanup-health-inline

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands to manage the environment and verify code integrity, specifically using mkdir for directory creation and pnpm for running type checks and build scripts.
  • [DATA_EXFILTRATION]: The workflow includes git push and bd sync operations. These commands synchronize local code changes and issue management metadata with remote source control and the Beads service, which is consistent with the skill's stated maintenance purpose.
  • [REMOTE_CODE_EXECUTION]: The skill orchestrates codebase analysis and modification by invoking subagents (dead-code-hunter and dead-code-remover) via the Task tool. These operations are performed within the agent's execution context to automate development tasks.
  • [PROMPT_INJECTION]: The instructions provided to subagents were reviewed for malicious patterns; they are focused on functional code analysis and cleanup, with no attempts to bypass safety filters or override system constraints detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 01:47 PM
Security Audit — agent-trust-hub — cleanup-health-inline