things-todo

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the things CLI utility from a public GitHub repository (github.com/ossianhempel/things3-cli) using the Go package manager.
  • [COMMAND_EXECUTION]: The skill relies on shell command execution to interact with the local operating system, including installation, environment sourcing, and binary execution.
  • [DATA_EXFILTRATION]: The skill accesses the local ~/.profile file to retrieve a THINGS_AUTH_TOKEN. It includes explicit safeguards, instructing the agent never to print the token and to redact it in logs when running dry-run operations.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from the user's Things 3 database (task titles, notes, and tags). Malicious instructions embedded in these fields could attempt to influence the agent's logic, though the skill provides no direct path for these instructions to escalate privileges.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 10:53 AM
Security Audit — agent-trust-hub — things-todo