debugging-local-replay

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous shell commands intended for system diagnostics, such as lsof for port checking, ps and pgrep for process identification, and docker ps for container health monitoring. It also includes a kill command aimed at cleaning up specific orphaned processes identified during diagnosis.
  • [CREDENTIALS_UNSAFE]: A debugging command in references/common-failures.md includes a hardcoded database URL (postgres://posthog:posthog@localhost:5432/posthog). These are default local development credentials used for a specific software environment and are not considered sensitive secrets.
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to run a service manually using npx tsx to capture startup errors that might be swallowed by a background watcher process. This involves executing local project code for troubleshooting purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:33 PM
Security Audit — agent-trust-hub — debugging-local-replay