Observe Codex Turn

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

The description goes well beyond the annotations by documenting fallback behavior ('After Bridge process loss, falls back to persistent thread/read history'), the precise semantics of wait_ms ('one bounded event-driven wait only when the live turn is active'), and the caveat that a long interval without new commands is not evidence of a stall. It reveals useful runtime behavior such as 'inProgress' snapshots and live-state unreconstructability without contradicting the read-only/idempotent annotations.

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?

The description is long but front-loaded with the core action, followed by wait semantics, fallback behavior, and a supervision workflow. Every sentence carries actionable information, though there is minor redundancy with the schema's wait_ms explanation; the density is appropriate for the tool's complexity.

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?

Despite having no output schema, the description explains the categories of returned data ('Bridge runtime events, pending requests, and terminal output'), the fallback path, and how the agent should react after each wake/deadline. It could be more explicit about the exact shape of the returned 'events/state' snapshot, but it provides strong contextual coverage for a complex observation tool.

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?

The schema already fully describes all four parameters (100% coverage), including wait_ms's event-driven nature. The tool description adds meaningful nuance by restricting wait_ms to 'only when the live turn is active' and emphasizing 'one bounded' wait, which clarifies the operational contract beyond the schema text.

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 uses a specific verb ('Read') and precisely names the resources: 'Bridge runtime events, pending requests, and terminal output for a thread.' It clarifies the observational scope with 'bounded incremental sanitized' and clearly distinguishes this from the sibling turn-management tools by framing it as the supervision/observation primitive.

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?

The description explicitly states when to use it: 'When actively supervising an in-progress turn, use repeated bounded-wait observe calls until terminal unless the user explicitly pauses or stops.' It also provides exclusions ('it is not polling or stall detection') and warns against treating inactivity as a stall, while naming follow-up actions (steer, respond, interruption).

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