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

Annotations already provide readOnlyHint=false, openWorldHint=true, etc. Description adds valuable context: consumes credits, runs independently, isolated VM. No contradictions. Together, they give a clear behavioral picture.

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?

Well-structured with key information front-loaded. A couple of sentences could be trimmed, but overall efficient and organized. Earns its length.

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?

Covers main points: purpose, parallelism, follow-up tools, cost, prompt advice. No output schema, but mentions return of thread id. Adequate for a delegation tool with many optional parameters.

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 baseline 3. Description adds some value by emphasizing prompt quality and noting default model behavior, but doesn't significantly expand on schema descriptions.

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 clearly states it starts a new Capy agent thread to perform engineering work in an isolated VM. It distinguishes from siblings by mentioning capy_wait and capy_status for follow-up, and explains that multiple calls yield genuine parallelism.

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?

Provides explicit guidance: 'Call this once per independent piece of work' and indicates alternative tools (capy_wait, capy_status) for monitoring. Also covers cost and prompt crafting advice. Could be stronger on when-not-to-use, but sufficient.

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

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

Adds behavioral context beyond annotations: large diffs are truncated, use stats_only to preview. Annotations already declare readOnlyHint and openWorldHint, so description supplements without contradiction.

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?

Two focused paragraphs. First sentence immediately states purpose. No wasted words; each sentence adds value.

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?

No output schema, but description hints at return (diff, file list) and covers truncation. Missing details on exact format, but acceptable for a diff tool with sibling context.

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 baseline is 3. Description adds context for task_id source and stats_only usage but does not significantly enhance parameter meaning beyond what the schema provides.

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?

Clearly states it gets the diff a Capy task produced for reviewing work. Distinguishes from siblings like capy_status and capy_list by focusing on code diffs.

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?

Explicitly mentions task ids come from capy_status, capy_wait, or capy_list, and suggests using stats_only for large diffs. Provides clear context but no 'when not to use' exclusions.

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

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

The description is consistent with annotations (readOnlyHint=true, openWorldHint=true) and adds behavior like ordering (newest first) and observable states. No contradictions.

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?

Two sentences, front-loaded with the core purpose, no filler. Every sentence is informative.

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 read-only list tool with 8 optional parameters and no output schema, the description covers key aspects: purpose, usage, ordering, and filters. Lacks mention of pagination via cursor, but not critical.

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% with well-described parameters. The description adds value by grouping filters (status, tag, PR state, query) and hinting at their use, going beyond the schema.

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 clearly states it lists agent threads in a project, newest first. It distinguishes itself from siblings like capy_status (single thread) and capy_projects (list projects) by specifying the resource and scope.

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 provides explicit usage context: 'Use this to see the whole fleet at once' and lists observable states. However, it does not mention alternatives or when not to use it, missing some guidance.

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

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

Annotations indicate the tool is not read-only (readOnlyHint=false) and not destructive (destructiveHint=false). The description adds that it sends a follow-up and explains modes, but does not disclose return behavior, whether it blocks, or how to monitor progress. It mentions opening a PR as a side effect, which adds behavioral context. No contradictions with 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 two paragraphs: first defines purpose and key use case (open PR), second explains the two modes. It is concise and front-loads the most important information. Could be slightly more structured (e.g., bullet points), but no unnecessary sentences.

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?

With 6 parameters and no output schema, the description covers the main purpose and delivery modes. However, it lacks explanation of return behavior, asynchronous nature, or how to combine with sibling tools like capy_wait for completion. The PR opening mention adds context but leaves details unspecified.

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 baseline is 3. The description adds value by explaining the 'mode' parameter's effect (interrupt vs queue) and the workaround for opening PRs. It does not elaborate on other parameters like model, reasoning, or build_model beyond what the schema provides, but the schema already describes them well.

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 clearly states the tool sends a follow-up instruction to an existing Capy thread and enumerates specific use cases (steer, answer questions, correct approach, request more work, open PR). This distinguishes it from sibling tools like capy_open_pr (which has no endpoint, so this tool is the workaround) and capy_status or capy_list.

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 provides explicit guidance on when to use the tool (sending follow-up instructions, including opening PRs) and explains the two delivery modes ('interrupt' vs 'queue') with their appropriate contexts. It does not explicitly compare to siblings like capy_stop or capy_delegate, but the context is clear enough for an agent to infer usage boundaries.

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

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

Annotations already indicate readOnlyHint=true and openWorldHint=true. The description adds value by specifying that the tool returns conversation history including user questions and agent responses, which aligns with annotations without contradiction.

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?

Two sentences, no wasted words. The purpose is stated upfront, and the description is efficiently structured.

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?

Given the read-only nature and schema coverage, the description adequately explains what the tool returns. No missing critical details, though pagination behavior is implied via the cursor parameter.

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 description does not need to explain parameters. It adds no extra meaning beyond what the schema provides, meeting the baseline expectation.

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 clearly states the verb 'Read' and the resource 'Capy thread's conversation', specifying it returns what the user asked and agent responses. It distinguishes from siblings like capy_message (sending messages) and capy_list (listing threads).

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 provides clear context for when to use the tool: to get the agent's summary, reasoning, or pending question. It does not explicitly exclude other scenarios or mention alternatives, but the context is sufficient.

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

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

Annotations already declare readOnlyHint and openWorldHint; description adds context about output content (provider and captain info) but no additional behavioral traits.

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?

Two concise sentences with no wasted words; purpose is front-loaded.

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 zero-parameter read-only tool, the description adequately covers what the tool returns and why it's useful.

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?

No parameters exist, so baseline is 4. Description doesn't need to explain any parameters.

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 clearly states it lists available models with their provider and captain capability, which distinguishes it from sibling tools like capy_status or capy_projects.

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?

Provides explicit usage scenarios (spreading parallel work, picking cheaper models) but doesn't mention when not to use it or compare to alternatives directly.

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

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

Annotations are present and the description adds context about asynchronous behavior (PR not existing on return) and the need to poll. No contradictions with annotations; useful behavioral disclosure beyond read/safety hints.

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?

Three sentences: first states the action, second explains mechanism, third gives post-call guidance. Front-loaded and no wasted words.

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?

No output schema, so description should cover return value. It mentions 'report any PRs on the thread' and polling, but does not specify what the immediate response contains (e.g., thread id or status). Adequate but has a gap.

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% with parameter descriptions. The description adds example values and clarifies that instructions can include title, base branch, or draft status, providing additional meaning beyond schema.

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 clearly states the tool asks an agent in a thread to commit work and open a pull request, using specific verbs and resource. It distinguishes from siblings like capy_message and capy_wait.

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 explains it is a convenience wrapper around capy_message and that the PR is not created immediately, advising to call capy_wait or capy_status afterwards. It lacks explicit when-not-to-use scenarios but provides clear usage context.

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

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

Annotations already provide readOnlyHint=true and openWorldHint=true. The description adds value by specifying the output (ids and connected repos) and clarifying that the list is limited to reachable projects. No contradictions.

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?

Two sentences: first states purpose and output, second provides usage guidance. No wasted words, front-loaded with key information.

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 simple list tool with well-documented parameters and annotations indicating read-only and open-world, the description covers the essential context. Could mention pagination behavior, but not critical.

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 covers 100% of parameters with descriptions (limit with defaults, cursor for pagination). The description does not add any extra meaning beyond the schema, so baseline 3 is appropriate.

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 clearly states the action ('List'), resource ('Capy projects'), and scope ('this API token can reach'). It also specifies return fields ('ids and connected repos') and provides a usage hint that distinguishes it from sibling tools like capy_delegate.

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 says 'Call this first if you do not know which project to delegate into,' giving a clear when-to-use directive and implying an alternative (delegate).

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

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

Annotations already declare readOnlyHint=true and openWorldHint=true. Description adds what is included (tasks, PRs) and confirms immediate return, aligning with 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?

Three sentences, front-loaded with purpose, no wasted words. Every sentence adds value.

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?

No output schema, but description explains what is returned (state, waiting, tasks, PRs). Mentions alternative tool. Complete for a one-shot status check.

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 baseline is 3. Description adds minor value by mentioning verbose includes 'tags, Slack links, and timestamps', but does not significantly expand on schema.

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?

Description clearly states verb (get), resource (thread state), and specifics (running, waiting, tasks, PRs). Distinguishes from sibling capy_wait by noting one-shot nature.

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?

Explicitly says when to use (one-shot check) and when not (use capy_wait to block). Provides clear context for choosing between tools.

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

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

Annotations already indicate destructiveness and idempotency. Description adds that committed work remains, clarifying the scope of stopping. No contradictions or additional details on permissions or side effects.

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?

Two concise sentences, each adding value: the first states the action, the second provides usage context and a behavioral caveat.

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 simple 1-param tool with good annotations, the description covers purpose, when to use, and key limitation. Could mention post-stop state or error handling, but sufficiently complete.

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% for the single parameter, so baseline 3 applies. The description does not add meaning beyond the schema's own description of thread_id.

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 clearly uses a specific verb 'Stop' and resource 'Capy agent thread', distinguishing it from sibling tools like capy_list or capy_status.

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?

Provides clear guidance on when to use ('cancel work going wrong', 'stop burning credits') and what it does not do ('work committed not undone'), but does not mention alternatives like checking status first.

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

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

The description adds valuable behavioral details beyond the annotations (readOnlyHint, openWorldHint): it explains the output structure (split by LLM/VM, user/thread), mentions the org_id limitation, and states default values for parameters. No contradiction with 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 two short paragraphs: the first states the core purpose, the second adds a necessary instruction about org_id. Every sentence contributes value with no redundancy, and the key information is front-loaded.

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?

The tool has 8 parameters (1 required) and no output schema. The description explains the purpose well, mentions default behaviors, and covers the most important parameter (org_id). It lacks explicit output structure details but describes the breakdown sufficiently for a reporting 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?

Although schema coverage is 100%, the description adds critical pragmatic information: default values for to ('now'), from ('7 days ago'), page (1), page_size (20), and routed ('paid'), plus context for org_id. This aids correct invocation beyond what the schema alone provides.

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 ('Report') and resource ('what your Capy agents have cost'), clearly distinguishing this from sibling tools like capy_status or capy_projects. It explicitly states the tool's function of reporting cost breakdowns by user and thread.

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?

It provides a concrete use case ('answer what did that fleet cost me') and explains where to obtain the required org_id (from the Capy web dashboard). While it doesn't explicitly mention when not to use it or alternatives, the context from sibling tools makes it clear this is the only cost-related tool.

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

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

Annotations include readOnlyHint, openWorldHint, idempotentHint. The description adds significant behavioral details: polling with backoff due to no streaming, early return on human blocking without error, and timeout behavior that allows retries. No contradiction with 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 concise, well-structured in three paragraphs, each sentence adding value. No redundant or vague statements. Front-loaded with the core purpose.

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?

The description covers all necessary aspects: blocking behavior, polling mechanism, edge cases (human block, timeout), retry advice. Despite no output schema, the description's claim to 'report what happened' is sufficient for the tool's simplicity.

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?

All three parameters have schema descriptions (100% coverage). The description reinforces purpose but does not add new meaning beyond what the schema provides. Baseline 3 is appropriate.

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 clearly states that the tool blocks until Capy threads finish and reports results. It specifies the action (block/wait), resource (Capy threads), and distinguishes it from siblings like capy_delegate (which starts agents) and capy_status (non-blocking check).

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 explains when to use: after delegating work via capy_delegate, and that it transforms background agents into a synchronous flow. It notes early return on human block and timeout retry, but does not explicitly compare to alternatives like capy_status for non-blocking checks.

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