Twin Me Up

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

Annotations already declare readOnlyHint=false, destructiveHint=false and idempotentHint=false, so the agent knows this mutates state harmlessly. The word 'merge' usefully signals additive/partial-update semantics rather than a full overwrite, but the description says nothing about permissions, error behavior on a missing episode_id, or how conflicts in merged fields are resolved.

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?

A single well-formed sentence with the action and the destination front-loaded and no filler. It is efficient, though it is arguably too terse to carry the behavioral context a mutating tool needs.

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 and a nested 'changes' object, the description covers the core intent but omits what the amended decision looks like afterward, whether the prior state is preserved, and the relationship to twin_record_decision. Adequate as a minimum but with clear gaps.

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 67% and the schema already documents 'changes' (merge semantics against record_decision names) and 'request_id' (idempotency key). The description adds the framing that changes are the principal's follow-up answers, but leaves episode_id entirely unexplained beyond its name.

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 ('merge') and resource (recorded decision) plus the source of the content (the principal's answers to follow-up questions). This distinguishes it from twin_record_decision (which creates the decision) and twin_record_correction, though it does not name those siblings explicitly.

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 mention of 'follow-up questions' implies the usage context (amending an already-recorded decision with answers gathered from twin_next_questions), but there is no explicit when-to-use/when-not-to-use guidance or reference to the alternative tools twin_record_decision and twin_record_correction.

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, idempotentHint=true and destructiveHint=false, so "it changes nothing" largely restates the structured data. The polling guidance adds a useful behavioral cue (safe to call repeatedly while blocked) but nothing about latency, expiry of approvals, or error cases is 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?

One sentence, front-loaded with the returned status values, followed by the polling directive. No filler and nothing repeated from the title.

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 no output schema, the description carries the return-value burden and does so adequately by enumerating the decision states and the unused flag. Combined with full parameter coverage and rich annotations, an agent has everything needed to call and interpret this 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?

There is a single parameter with 100% schema description coverage, and the schema itself explains that approval_id comes from twin_request_approval. The description adds no syntax, format, or validity constraints beyond that, 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?

The description states precisely what the tool returns: the principal's decision state (approved/declined/undecided) plus whether the approval is still unused. That makes it clearly distinct from siblings like twin_request_approval or twin_authorise, which act on requests rather than reading their status, though no sibling is named explicitly.

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?

"Poll this while you wait; it changes nothing" gives explicit usage context: this is the tool to call repeatedly while an approval is pending, and it is side-effect free. It stops short of naming an alternative for the case where the approval has already resolved, but the context is clear.

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 cover the safety profile (readOnly, idempotent, non-destructive), so the bar is lower; the description adds the decision semantics and the required next action, which is genuinely useful context. It does not clarify whether a third outcome (e.g. a hard denial) can occur, which matters for a gate on consequential actions.

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, both earning their place, with the trigger condition front-loaded before the outcome semantics. No filler or restatement of the title.

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 no output schema, the description carries the return-value burden and does explain 'allow' and 'ask'. It leaves open whether other outcomes exist and never explains the required 'workflow' parameter, both of which an agent gating real actions would want.

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%: action and attempt_id are documented in the schema, while workflow has no description in either place. The description adds no parameter-level detail (no action naming convention, no workflow semantics), so it does not compensate for the gap — 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 defines a specific function — a pre-flight authority check that returns 'allow' or 'ask' — and spells out what each outcome means. It is clearly distinguishable from the sibling twin_request_approval, which it names as the follow-up path.

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 states exactly when to call it ('before any consequential action: sending, paying, committing, writing to live systems') and what to do on each result, routing explicitly to twin_request_approval for the 'ask' case. Nothing is left to inference.

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 non-destructive, so the safety profile is covered. Beyond that, the description discloses the content of the brief (limits, authority, stop conditions, reporting expectations) and an instruction to 'embed and follow it', which is genuine behavioral context. It stops short of describing output format or failure modes.

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 tight sentences, front-loaded with the timing trigger and then the payload. Every clause carries information; there is no filler.

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 no output schema, the description usefully enumerates what the brief contains (guidance, limits, authority, stop conditions, reporting) and the timing trigger. It is nearly complete, though it could note idempotent re-calling or how objective/actions shape the returned guidance.

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 the schema already documents all four parameters with format, pattern and examples. The description adds no parameter-level 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?

The description states a specific verb and outcome: 'Returns the principal's confirmed guidance, hard limits, your authority, stop conditions and how to report back.' This tells an agent exactly what the tool yields. It does not explicitly contrast itself with siblings like twin_track_record or twin_next_questions, so it falls short of a 5.

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?

'Call at the start of work in a workflow' gives a clear triggering condition. However, it names no alternatives and states no when-not-to-use case, leaving the agent to infer that later-stage siblings supersede it.

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, destructiveHint=false, so the safety profile is covered. The description adds the useful filtering context that only unanswered, high-value questions are returned. It does not disclose behavior around invalid episode_id or how many results come back when limit is omitted.

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 short sentences, front-loaded with the core scope and ending with the workflow nudge. Nothing is wasted, though the brevity borders on under-specification rather than tightness.

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 no output schema, no parameter documentation in the schema, and an ambiguous episode_id, the description should carry more of the burden. It omits result structure, limit behavior, and explicit linkage to the amendment tool, leaving an agent to infer too much.

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% and there are two parameters. The description never explains what episode_id identifies or what limit controls (default 3, max 10), so it fails to compensate for the missing schema documentation.

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 identifies the resource (follow-up questions about a recorded decision) and implies a specific scope ('most valuable unanswered'), but it leads with a noun phrase rather than a verb and never states the operation outright. It is distinguishable from most siblings, though it relies on the title for the 'get' framing.

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?

'Ask them, then amend' implies a workflow that leads into a decision amendment, which is useful implied guidance. However, it never names the sibling tool (twin_amend_decision) nor states conditions such as when to call this versus twin_brief or twin_review_queue.

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 this is a non-read-only, non-destructive write, but the description adds genuinely useful lifecycle context that annotations cannot convey: the lesson 'is never authoritative until confirmed', meaning the write is provisional. It stops short of explaining what confirmation involves or how the pending state is surfaced, but the core behavioral caveat is 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?

Two short sentences with zero padding; the provisional nature of the tool is front-loaded in the first clause. Nothing in the text 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?

For a nine-parameter write tool with low schema coverage and no output schema, the description is thin: it covers the tool's role and lifecycle state but says nothing about the workflow/kind taxonomy, how lessons are reviewed or confirmed, or how to check status (e.g., twin_approval_status). Adequate for orientation, incomplete for correct invocation.

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 only 33%, so several of the nine parameters (kind, workflow, conditions, confidence, episode_ids, review_by) are undocumented in both schema and description. The description adds no parameter meaning whatsoever, leaving the agent to guess at fields like kind, workflow format, and conditions.

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 ('Propose a lesson') plus the recipient ('for the principal to confirm'), so the agent knows this creates a provisional lesson rather than a settled fact. It does not explicitly distinguish itself from siblings like twin_record_decision or twin_record_correction, so sibling differentiation is left to inference.

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: the phrase 'for the principal to confirm' suggests this is the pre-confirmation path, but the description never says when to choose this over twin_record_decision, twin_record_correction, or twin_request_approval, and offers no prerequisites or 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?

Annotations already declare write semantics (readOnlyHint=false, destructiveHint=false), so the safety profile is covered. The description adds that the record becomes "learning material," implying an append-only store, but says nothing about immutability, whether corrections can be edited later, or how retries behave beyond the schema's request_id note.

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 short sentences, with the triggering condition front-loaded ahead of the storage behavior. Every clause earns its place and no filler is present.

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 seven-parameter write tool with no output schema, the description covers the core intent but omits practical guidance: whether entries are immutable, how to avoid duplicate records (beyond the schema note), and when to prefer a sibling tool such as twin_record_decision. Adequate but with clear gaps.

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 71% and the description touches the key payload fields implicitly via "both versions and the reason." It adds no guidance on workflow, situation, sources, or request_id, so it neither compensates for nor exceeds what the schema already documents. 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 (capture/record) and resource (a correction from the principal), and clarifies that both the agent's version and the human's decision are stored. It is distinguishable from siblings like twin_record_decision, though it does not explicitly draw that line.

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?

"Call whenever the principal edits, overrides or rejects your work" gives an explicit trigger condition, which is strong usage guidance. What is missing is any exclusion or routing rule versus adjacent tools (twin_record_decision, twin_amend_decision, twin_propose_lesson).

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 the safety profile (write, non-destructive, not idempotent, not open-world), so the description's burden is lighter. It adds one useful behavioral fact beyond annotations — 'Returns follow-up questions' — but says nothing about what recording does to downstream state, whether escalation triggers fire anything, or how duplicates are handled.

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?

One dense sentence plus a short return-value fragment, front-loaded with the core verb and the field inventory. Every clause maps to a real parameter, though the comma list is compressed enough that an agent must parse it carefully.

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 14-parameter mutation tool with 21% schema coverage and no output schema, the description is thin: it covers roughly the decision narrative but omits required-parameter emphasis (workflow and situation are required), provenance via sources, outcome, missing_information, and idempotency. The single-line return note does not compensate for the missing parameter guidance.

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 only 21% across 14 parameters, so the description has to carry weight; it does map several natural-language phrases to parameters (situation, options/choice, reasons for rejected options, cues, 'what would change it' → reversal_conditions, 'when to escalate' → escalation_triggers). However, it leaves workflow, goal, rationale, sources, outcome, and missing_information unexplained, and gives no format or coupling guidance.

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+resource ('Capture a real decision') and enumerates the substance being captured (situation, options, choice, rejected-option reasons, cues, escalation triggers), so an agent knows exactly what payload this tool wants. It does not, however, explicitly differentiate itself from twin_amend_decision or twin_record_correction, so it falls short of a 5.

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 word 'real' hints that this records actual (not hypothetical) decisions, but there is no stated when-to-use, no prerequisite, and no routing against the 11 siblings such as twin_amend_decision (existing decision) or twin_record_correction. An agent must infer that this is the creation path.

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 declare this is a mutation (readOnlyHint=false, destructiveHint=false), and the description adds genuinely new behavioral facts: the principal's decision happens in their own terminal, the caller must not act before approval, and approval is scoped to a single attempt unless extended. It does not say what a decline does or what is returned, but the one-attempt scoping is information the annotations cannot 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 short sentences, zero filler, and the trigger condition is front-loaded before the constraint and the post-condition. Every clause carries actionable 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 6-parameter mutation tool with no output schema, the description covers the trigger, the gating rule, and the approval scope, which is enough to call it correctly. It leaves the decline path and the shape of the recorded request unstated, minor gaps given the sibling status tool covers outcomes.

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 67%, so most parameters carry their own documentation (action example 'repo.push', details, summary, sources, request_id). The description adds almost nothing parameter-level beyond reiterating that the action/work must be approved. 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 states a specific verb and resource: it records one approval request that the principal approves or declines in their own terminal. It also implicitly separates itself from sibling twin_authorise (which decides 'ask') and twin_approval_status (which reports the outcome), so an agent can place it in the workflow without guessing.

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 an explicit trigger condition — 'Call when twin_authorise says ask and the work needs that action' — and names the follow-up gate ('Do not act until twin_approval_status says approved'). This is a precise when-to-use rule with the surrounding siblings named, rather than inferred usage.

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, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds that the queue contains unconfirmed lessons for confirmation, revision, or rejection, but omits return format, pagination, and authentication context.

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?

It is a single, short phrase with no wasted words. It is front-loaded with the resource, though it is a sentence fragment rather than a structured description.

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 read-only queue with no output schema, the description gives the essential return content ('unconfirmed lessons'). However, it does not compensate for the undocumented workflow parameter and provides no additional behavioral context beyond what annotations already cover.

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 one optional parameter, 'workflow', with 0% description coverage, and the description does not mention it at all. An agent gets no explanation of what the workflow filter does or how omitting it behaves.

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 identifies a clear resource and state: 'Unconfirmed lessons' awaiting the principal. It does not state an explicit verb like 'list' and does not differentiate itself from siblings such as twin_approval_status or twin_search, but the purpose is still clear from the noun phrase.

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 phrase 'to present to the principal' implies the use case—gathering items before a principal review—but there is no explicit when-to-use guidance, prerequisites, or comparison to alternative sibling tools. Usage is inferable rather than stated.

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, destructiveHint=false and closed-world scope, so the safety profile is covered. The description adds only the constraint that results are restricted to confirmed (not draft/proposed) guidance, which is genuine but modest additional context.

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?

A single tight sentence with no filler, and the core action is front-loaded. It is arguably too terse given the gaps, but nothing is wasted.

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 carries the burden of describing the response, yet it says nothing about what a match looks like, ordering, relevance ranking, or how `limit` behaves. For a search tool with two undocumented parameters, this leaves the agent under-informed.

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 only 33% – only `query` is documented in the schema. The phrase "within a workflow" hints at the `workflow` parameter's role but never explains its identifier format or purpose, and the `limit` parameter is left entirely unexplained in both schema and description.

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 ("Find") and a well-defined resource ("the principal's confirmed guidance") scoped to a question and a workflow. The word "confirmed" usefully distinguishes this from adjacent concepts like proposed lessons, though no sibling tool is named explicitly.

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 implies the context of use (looking up guidance for a question inside a workflow) but gives no when-to-use vs when-not guidance and never references an alternative such as twin_brief, twin_next_questions, or twin_review_queue. The agent must guess which of the twelve twin_* tools to reach for.

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 readOnly, idempotent, non-destructive and closed-world, but the description adds real value by resolving the ambiguity created by the name 'track_record' – the tool only surfaces evidence and recommends, it does not promote or mutate state. It still says nothing about cost, latency, or how much history is scanned.

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 compact sentences that front-load what the tool returns before the caveat about promotion. The list is telegraphic but dense; nothing is padding, though the compressed noun stack takes a moment to parse.

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 is the only signal about what comes back; it lists the evidence categories and the recommended next step, which is helpful. It nevertheless leaves the stage vocabulary and workflow identifier unexplained for a fairly abstract learning-loop 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?

There is one required parameter with 0% schema description coverage, so the description must carry the burden of explaining what a 'workflow' identifier is and what form it takes. It only alludes to 'a workflow' generically and adds no format, naming, or lookup guidance beyond the regex pattern already 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?

States a specific resource (evidence and stage for a workflow) and enumerates the kinds of evidence returned: confirmed learning, corrections, evaluation results, grants, and recommended next step. It partially differentiates from siblings by declaring it only recommends, unlike the promotion/authorise tools, but it never names 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?

Usage is implied rather than stated – an agent infers you call this to inspect how well the twin knows a workflow. The closing clause 'it recommends; only the principal promotes' hints that a different tool handles promotion, but no alternative is named and no when-not condition is given.

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