principle-laziness-protocol
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local commands using
codex execand thespawn-codex-workerscript for task delegation. These are presented as standard operational procedures for the agent within its host environment. - [PROMPT_INJECTION]: The skill defines behavioral overrides and model selection policies, which guide the agent's internal reasoning and delegation logic.
- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection via the processing of untrusted subagent outputs and task calls.
- Ingestion points: Task or subagent calls as described in SKILL.md.
- Boundary markers: None present in the instructions.
- Capability inventory: Execution of
codex exec,spawn-codex-worker, host terminal tools, and git utilities (SKILL.md). - Sanitization: No specific sanitization, validation, or escaping of input data is mentioned.
Audit Metadata