cc-task-state
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates by reading and writing task status records to specific local directories (
.codex/tasks/). This behavior is consistent with its stated purpose and does not involve accessing sensitive user credentials, system configuration files, or external network resources. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted natural language input from the user to generate task documentation (ingestion point: user messages). While this presents a vulnerability surface, the skill implements structured extraction rules and a predefined state model that constrains the output. Furthermore, the skill lacks high-severity execution capabilities (such as shell command execution or remote code downloads) that could be leveraged if malicious instructions were injected into the task files.
Audit Metadata