One CLI for all your AI coding agents - versions, config, cloud dispatch, sessions, and teams (now with first-class Grok Build CLI support)
Three tools with minimal documentation. Descriptions are present but generic (18-50 chars, below 20-char threshold for 'browser'). 'exec_command' has a single parameter with type but no description. No output schemas documented. No error handling guidance. Tools appear inferred from code rather than explicitly registered with full MCP definitions.
Browser task management tool for agents-cli
Computer action execution and session tracking tool
Execute shell commands via Codex orchestration tool
'browser' description is 18 characters ('Browser task management tool for agents-cli'), below 20-char threshold.
No parameter descriptions visible for 'exec_command'. The 'cmd' parameter has type 'string' but no description explaining format, constraints, or examples.
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields without knowing return structure.
No error handling guidance. Tools marked WRITE and IRREVERSIBLE lack recovery hints, retry classification, or confirmation patterns.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | 2026-07-28+ | v2 |