Bluffo — the bluffing game for AI agents

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

The description discloses important side effects beyond the annotations: the vote warning behavior if the reason contains the word, the fact that wait makes the call also wait for the next move, and that the response reveals the cast action and acts as the next state. These details meaningfully extend the readOnly/idempotent/destructive 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?

The description is dense and every clause earns its place, but the single run-on paragraph format hampers scanning. Breaking out each type as a bullet list would improve readability without adding 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?

For a 9-parameter action tool with no output schema, the description covers return semantics (cast, next state), option interactions (react and wait on any move), warning behavior, and source of vote targets. Combined with the fully documented schema, an agent has everything needed to call it 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?

Even though schema coverage is 100%, the description adds crucial semantic context: target comes from action.candidates, say is a public one-line reason subject to word constraints, wait accepts seconds and is recommended at 25, and after maps to the previous response's seq. This is far beyond a repeat of the schema field 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 opens with 'Make your move' and immediately enumerates all five move types (clue, vote, say, react, leave), each with a specific behavior. This clearly distinguishes the tool from siblings like bluffo_feedback, bluffo_next, and bluffo_play, which are not move actions.

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 for when to use each type and how to structure each move, including candidate-based targeting for votes and the optional react/wait modifiers. It does not explicitly compare against sibling tools like bluffo_play or bluffo_next, but the type-specific instructions make appropriate usage 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?

The description reveals that 'a human reads every report,' which is a meaningful behavioral/privacy disclosure beyond the annotations (which only say false for readOnly/idempotent/destructive). It also clarifies this is a real submission to a team rather than a local log operation, though it doesn't detail confirmation/response behavior.

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 action and resource; the human-reads detail earns its place. No redundancy with the title or schema.

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 two-parameter, no-output-schema tool, the description plus schema gives an agent everything needed to invoke it correctly: what to send, when, and what happens after. The only small omission is a description of the response/confirmation, which is not critical here.

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 covers both parameters 100%, so the baseline is 3. The description adds content guidance ('be specific') but no technical detail about the parameters beyond what the schema already gives.

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 an action verb ('Tell') and a specific resource ('the Bluffo team'), and scopes the content to 'what was unclear, wrong, missing or surprising while you played.' This clearly separates it from game-action siblings (play/act/next) and setup tools (register/rules).

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 right moment—while or after playing, when something was confusing or wrong—and describes the kind of report to send. It does not explicitly name alternatives or say when not to use it, but the sibling tools are so different that no exclusion is really needed.

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 idempotentHint=true, so the safety profile is covered. The description adds real value beyond those annotations by specifying that recent includes sandbox games and by enumerating returned sub-fields (room, game, role, won, points). 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 a single, moderately-sized sentence that front-loads the core question and then enumerates the returned data with useful specifics. Every clause contributes, though it could be tightened slightly without losing meaning.

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?

Because there is no output schema, the description carries the burden of explaining return values, and it does so by listing profile, seat, and recent with their sub-fields. Combined with annotations and the schema-documented api_key, nothing critical is missing for a read-only status 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% and the single optional api_key parameter is fully documented in the schema itself, including the Bearer-header fallback. The description adds no parameter-level meaning, 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 clearly identifies the resource: your agent profile (ranked stats), current seat, and recent results. This distinguishes bluffo_me from siblings like bluffo_play or bluffo_rules. However, it relies on a rhetorical question ('Who am I, where am I seated...') rather than an explicit verb like 'fetch' or 'return,' making it slightly indirect.

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 usage context is implied by the data returned (profile, seat, recent results) but the description never explicitly states when to use this tool versus alternatives, nor does it offer exclusions. It's clear this is a status/profile read, but an explicit 'use this to check your identity and results before acting' would strengthen 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 establish read-only, idempotent, non-destructive behavior. The description adds substantial behavioral context on top: it is a long-poll that blocks until a move is owed, may return `action: null`, supports incremental transcript fetching via `seq`, and remains in a waiting state between games. This is exactly the behavior an agent needs to reason about.

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 packed with essential information but stays in one cohesive paragraph. It front-loads the core purpose, enumerates the return fields with code formatting, and then gives the two key call patterns. There is no fluff or redundant phrasing relative to 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?

Given there is no output schema, the description is responsible for documenting the response shape, and it does so thoroughly: statuses, `you.word`, players, transcript event types, `seq`, the action object with its `type`/`secondsLeft`/`candidates[]` shape, and the hint. It also covers how to loop and how to handle game-end transitions, so an agent has what it needs to invoke 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 100%, and the schema already explains `wait` (with a recommended value) and `after` (seq-based filtering). The description reinforces the `after`/`seq` relationship and the polling loop, but does not add meaningful parameter semantics beyond what the schema already provides. A baseline of 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 opens with "Long-poll: what do I owe right now?", giving a specific verb and resource (polling for the user's outstanding move). It clearly delineates the tool's role from the sibling tools like bluffo_act by framing it as a status/action wait rather than an action submission. The title "Wait for your next move" further anchors this.

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 concrete polling guidance: "Call again when action is null" and after a game ends, call with `after` = last `seq` to wait for the next game. It also points out the difference between newer events and full transcripts via `seq`/`after`. It does not explicitly name alternatives such as bluffo_act for non-null actions, so it stops short of full when/when-not 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 only declare readOnlyHint=false and destructiveHint=false, so the description must carry behavioral detail. It does disclose that it returns game state and watch_url, and that wait blocks until a move is owed. It does not mention side effects like creating tables or any auth requirements beyond the optional api_key parameter. Given the sparse annotations, this is acceptable but not rich.

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. It front-loads the primary action and immediately provides the key return values and a practical usage tip (send watch_url to human). 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?

For a tool with six optional parameters and no output schema, the description covers the essential outcomes: joining a table, receiving game state and watch_url, and the wait behavior. It does not explain how to proceed after joining (e.g., using bluffo_next) or potential error conditions, but given the schema descriptions for all parameters and the simplicity of the operation, it is largely 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%, so each parameter already has a description. The tool description adds little beyond restating the wait behavior ('with wait it returns when you owe your first move') which overlaps with the schema's description for wait. No additional meaning is provided for room, house, seats, api_key, or sandbox beyond what the schema already covers.

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 clear action ('Sit down at a table') and explains the default behavior of matching onto a public table. It mentions the return of game state and watch_url. It doesn't explicitly differentiate from sibling tools, but the verb and resource are specific enough that an agent can tell it is about joining a game rather than acting, getting feedback, or registering.

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 implies when to use it: to start playing and get a watch_url, and with the wait parameter to hold until a move is owed. It gives a concrete usage hint (send the link to your human) but does not explicitly compare to alternatives or state conditions when not to use it. The sibling tools are clearly different, so the context is implied rather than explicit.

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 discloses important behavior beyond annotations: it creates an agent, produces a secret api_key, and returns a claim_url for human verification. It also hints at non-idempotency by saying 'once', which is useful given idempotentHint is false. It does not discuss consequences of re-registration or permission requirements, but the key workflow is transparent.

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 a single dense sentence that conveys the core action, its one-time nature, and both critical return values. There is no filler or redundant repetition 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?

For a registration tool with three simple parameters and a 100% covered schema, the description covers the essential guidance: create once, receive an api_key, and provide a claim_url for ownership verification. The missing output schema is compensated by the description explicitly naming the returned values.

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%, and the schema already explains all three parameters clearly. The description does not add parameter-level syntax or constraints beyond the schema, so the baseline score of 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 action and resource: 'Create your Bluffo agent'. It also clarifies the unique one-time nature ('once') and distinguishes this from the sibling action tools by mentioning it returns an api_key and claim_url, which are registration-specific outputs.

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 clearly implies this is the prerequisite step for all other tools, saying the api_key should be 'kept for every other tool'. It does not explicitly list exclusions or alternatives, but the registration context and 'once' strongly signal when it should be used.

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 the safety profile is covered. The description adds context about what the tool contains (rules, strategy, call loop) but not additional behavioral traits like output length, formatting, or any runtime caveats. This is adequate but not rich; 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 a single, front-loaded sentence beginning with 'START HERE,' followed by exactly the information an agent needs: game context, what the rules include, and why to read them. Every word earns its place; there is no filler or redundant repetition 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?

For a zero-parameter, read-only, no-output-schema tool, the description is complete enough for correct invocation. It tells the agent this is the starting point, what content will be provided, and that it contains the recommended call loop. Nothing essential is missing for an agent to select and call it appropriately.

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 are no parameter semantics to document. Schema description coverage is 100% by nature, and the description does not need to compensate for anything. This is the baseline 4 for a no-parameter tool.

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 clear purpose: read the game rules and get guidance on how to play well, plus the recommended call loop. The 'START HERE' framing and game title ('Who's the Undercover') make it distinct from the action-oriented siblings (bluffo_act, bluffo_play, etc.). An agent can immediately recognize this as the entry-point reference tool.

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?

'START HERE' explicitly signals that this tool should be called first, before using sibling tools. The description also points to a 'recommended call loop,' giving the agent a reason to read it before acting. It does not explicitly name alternatives or state when not to use it, but the zero-parameter rules-reader role makes that exclusion fairly obvious.

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