implement

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of disclosure — and it excels. It reveals this is a workspace-write operation in a sandbox, enumerates mutation side effects (edits repo, runs builds/tests, fixes breakage), and explains the output contract (git diff appended for review). It even flags idempotent-ish behavior ('reads .codex-bridge/ plan files if present') and cost implications. All side-effect domains an agent would need to know before invoking are disclosed.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Four sentences that front-load the core purpose and flow logically through prerequisites, mechanics, and caveats — every clause earns its place. The emphasis via ALL-CAPS 'DETAILED' and the cost warning are placed near the end where they're instructive, not distracting. Minor deduction for slight redundancy between 'Reads .codex-bridge/ plan files if present' and the later re-explanation of the task plan format; these could be tightened into one thought.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a high-stakes mutation tool with no annotations and no output schema, this description is admirably complete: side effects, prerequisites, failure semantics, cost, and output destination are all covered. It tells the agent what not to delegate to it ('never thinks or designs') and when to decline (if plan isn't written). Sibling tools like write_file, run_command, and git_commit make similar repo mutations, yet this description fully distinguishes the delegated-execution model without padding.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3 per the rubric — the schema already documents all four parameters, including practical notes like 'never guess' on project and 'new chat must be re-armed' on session_token. The description reinforces the 'task' semantics (detailed step-by-step, executed literally) which echoes the schema's 'Codex follows it literally' language, but adds no genuinely new parameter-level insight beyond what structured fields already state.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a clear verb+resource ('Run Codex...on the current task') and details the full scope of behavior: reads plan files, edits repo, runs builds/tests, fixes breakage, appends diff for review. It distinguishes itself from siblings by emphasizing execution over planning ('never thinks or designs') and explicitly warns 'Only call after ChatGPT has written that plan' — differentiating from plan_write/plan_read. No ambiguity about what this tool does or where it fits in the workflow.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit when-to-call guidance: 'The task must be a DETAILED step-by-step plan' and 'Only call after ChatGPT has written that plan.' It also signals when NOT to use it ('never thinks or designs'), establishes a hard prerequisite (plan must exist first), and notes the operational cost of invocation ('Costs Codex usage'). For an agent choosing between this and the planner siblings, the decision rule is crystal clear.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.