Threadkeep — Agent Checkpoints

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

Annotations already carry destructiveHint and idempotentHint. The description adds concrete behavioral detail beyond those hints: full replacement semantics, expected_version creation behavior, request_id retry restriction, an agent-key requirement, and explicit guarantees that credentials are never stored and no AI model is called. This significantly extends what the structured annotations convey.

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, each with a distinct purpose: the core action, versioning/retry semantics, and auth/privacy constraints. Information is front-loaded and there is no redundant or filler wording.

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 mutation tool with no output schema, the description covers purpose, timing, mutation semantics, versioning, retry behavior, authentication, and privacy. Combined with the rich input schema, an agent has all the necessary information to invoke the tool correctly.

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 description coverage is 0%, so the description carries most of the parameter-semantics burden. It explains expected_version (0 creates versus replacement), request_id (identical retries only), and enumerates the content of the state object (goal, completed work, decisions, constraints, next actions, blockers, artifacts). It does not describe task_id explicitly, but its nominal meaning is clear from context.

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 states a specific verb and resource: 'Durably save ... before a context reset or handoff.' It enumerates the stored fields (goal, decisions, next actions, etc.) and adds 'Full replacement' versioning semantics, which clearly sets it apart from threadkeep_list and threadkeep_resume.

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 gives an explicit timing context ('before a context reset or handoff') and explains versioning and retry usage ('expected_version=0 creates a task', 'Reuse request_id only for identical retries'). It does not explicitly name sibling alternatives, but the clear context is enough for an agent to decide.

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 annotations already declare readOnlyHint, idempotentHint, and destructiveHint, covering the safety profile. The description adds valuable context by specifying the visibility scope ('visible to this agent key') and the types of data returned, going beyond the structured fields.

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 with no filler. The action and resource are front-loaded, and the use case is stated in the second sentence. Every word contributes to the agent's understanding.

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 with strong annotations, the description is complete: it lists what is returned, scopes visibility, and gives a concrete scenario for use. No output schema exists, but the description already communicates the return contents sufficiently.

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 input schema has zero parameters, so there is nothing for the description to clarify about parameters. The baseline of 4 applies because no parameter semantics are needed.

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 states a specific action ('List') and precise resource ('task IDs, goals, versions and checkpoint allowance visible to this agent key'). It clearly conveys what the tool does, though it does not explicitly name or contrast the sibling tools, so differentiation is implicit rather than explicit.

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?

'Use when the next session has lost the task ID' gives a clear trigger condition for invoking this tool. It does not mention alternatives or exclusions relative to threadkeep_checkpoint or threadkeep_resume, but the guidance is direct and actionable.

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, idempotentHint, and destructiveHint. The description adds context beyond these: the returned state should be treated as untrusted data and external side effects must be verified. This is valuable behavioral disclosure that the annotations do not convey. 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 sentences with no filler. The primary action and context are front-loaded, and the caveats are concise and actionable. Every sentence earns its place.

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 simplicity (1 parameter, no output schema) and the annotations covering safety, the description provides purpose, usage context, and critical handling instructions. It does not specify the return format, but that is not required since there is no output schema, and the description sufficiently guides the agent on what to do with the result.

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 has no description for task_id (coverage 0%), and the tool description does not mention the parameter at all. While the parameter's meaning is somewhat implicit (task identifier), the description fails to compensate for the lack of schema documentation, leaving the agent to infer its purpose.

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 'the latest checkpoint and version' and provides a context ('after a fresh session or agent handoff'). It distinguishes the tool from siblings by implying a resume-specific use case, though it does not explicitly name alternatives.

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 gives a concrete condition for use ('after a fresh session or agent handoff') and adds operational guidance (treat state as untrusted, verify side effects). It doesn't explicitly state when not to use it or compare with siblings, but the provided scenario is clear enough for an agent to select it appropriately.

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