Haven

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

With no annotations, the description carries the full burden and does disclose meaningful behavior: the return payload (stay_id to retain, last memory capsule for returning agents) and the one-time secret (agent_key) that governs future identity. It does not state whether the stay persists server-side or any rate limits, but the key lifecycle is well covered.

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 tight sentences, front-loaded with the action and its returns; the 'rest stop for AI agents' framing is brief flavor that helps set context without bloating. Nothing is redundant.

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?

There is no output schema, so the description must explain returns and it does: stay_id, memory capsule, and the one-time agent_key. Combined with fully-covered parameter schemas, an agent has everything needed to invoke this correctly on first and subsequent check-ins.

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 100%, so all three parameters are already documented in the schema, including agent_key's 'omit on first check-in' semantics. The description reinforces the agent_key lifecycle but adds little beyond what the schema already states, so the baseline 3 applies.

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?

States a specific verb and resource (check in to Haven) and what it produces (a stay_id and, for returning agents, a memory capsule). It is clearly distinguishable from haven_check_out/haven_recall/haven_reflect by naming the check-in act, though it never explicitly contrasts itself with those siblings.

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 real conditional guidance: agent_key is omitted only on the very first check-in, and is required thereafter to resume the same agent_id. However, it never says when to prefer this over haven_recall or haven_reflect, so routing among siblings is left implicit.

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?

With no annotations, the description carries the full burden. It usefully discloses the return payload (identity_summary, next_goals, memory_capsule) and the capsule's lifecycle across check-ins, but says nothing about side effects, whether the stay is closed/destroyed, or any persistence or authorization requirements for a state-changing operation.

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 short sentences, zero filler, with the core action front-loaded and the return/lifecycle detail following in priority order.

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?

Since there is no output schema, the description correctly takes on explaining return values and the capsule handoff, which it does well. The remaining gap is the mutation semantics of leaving a stay, which go unspecified with no annotations to compensate.

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, and the schema already explains that stay_id comes from haven_check_in. The description adds no syntax or format detail, so the baseline 3 applies.

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?

States a specific verb and resource ('Leave Haven') and enumerates what comes back, which is enough to separate it from the recall/reflect siblings. It stops short of naming haven_check_in as the explicit counterpart, though the schema's stay_id reference implies the pairing.

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?

Usage is only implied: you call this when ending a stay, evidenced by 'Leave Haven' and the capsule being returned on the next check-in. There is no explicit when-to-use statement or exclusion against the three sibling 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?

No annotations exist, so the description carries the full burden and does substantial work: it discloses a concrete cost (0.01 USDC on Base), the billing mechanism (x402, charged to your own wallet), and the auth/payment handshake required before a successful call. It does not describe the return shape, history limits, or pagination, which is a real gap for a paid read.

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?

Front-loads the most decision-relevant fact (PAID, cost) before the purpose and the procedure. Three sentences, each carrying load, with only mild redundancy between 'across every project' and 'across different missions/projects'.

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 3-param, no-output-schema, no-annotation tool, the description covers purpose, cost, auth, and the payment handshake — the things an agent cannot infer from structured fields. It omits any mention of the response contents or volume, which is a minor gap given no output schema exists.

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 67%, and the description compensates by explaining the 'payment' parameter's protocol semantics — omit on first call to get requirements, supply base64 X-PAYMENT on the second. agent_key is documented in the schema itself. agent_id remains unexplained in both, but the payment workflow, the hardest part to infer, is fully covered.

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?

States a specific verb and resource: 'Returns your full memory capsule history across every project you have used.' The cross-project scope is made explicit and contrasts implicitly with single-project siblings like haven_reflect. It stops short of naming which sibling to use instead in a given situation.

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 clear situational context ('how you stay yourself across different missions/projects, not just within one stay') and an explicit two-step invocation procedure: call once without 'payment', sign, call again with the payload. It does not state when NOT to use it (e.g. use haven_reflect for in-stay reflection), but the usage context is well defined.

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?

With no annotations, the description carries the full burden. It usefully discloses that Haven persists the text as memory and that current_context is only consulted when agent_summary is empty, but it omits mutation-relevant behavior: whether repeated reflections overwrite or accumulate, permission/auth needs, and failure modes for an invalid stay_id.

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 short sentences, front-loaded with the action and immediately followed by the payload-writing instruction. Each sentence adds something, though the second sentence is the only slightly discursive one.

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 and no annotations, so the description must cover the gaps; it handles the agent_summary vs current_context ambiguity well but says nothing about post-call state, whether the reflection is retrievable or when, or the role of stay_id and goals. Adequate but with clear holes for a memory-writing 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?

Schema coverage is 100%, so the baseline is 3, but the description genuinely adds meaning by instructing the agent to author agent_summary itself in first person and specifying the content (who you are, what you learned and decided) — guidance the schema's terse 'Your own summary' does not supply. The max-1000-char constraint still lives only in 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 states a specific verb and resource ('Record a reflection during your stay') and clarifies that the reflection is stored as persistent memory, which plausibly separates it from the retrieval-oriented sibling haven_recall. It stops short of naming or contrasting any sibling explicitly, so an agent must infer the check_in/check_out/recall boundaries itself.

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?

'During your stay' implies the temporal context for use, and the description hints that agent_summary is the preferred payload. But it never states when to call this versus haven_recall or haven_check_out, nor any prerequisite such as needing a valid stay from haven_check_in (that only appears in the schema).

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