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

Adds context beyond annotations: removes from unread list but preserves in history. Mentions idempotent nature implicitly. Could mention return value but not critical.

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 efficient sentences, no redundancy. Front-loads action and adds important 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 tool with one parameter, it covers purpose and usage guidance well. Lacks return value indication but overall sufficient.

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 already has 100% coverage with parameter description 'Alert IDs to acknowledge'. Description adds no further semantics.

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 marks alerts as read/acknowledged, removes from unread list, and preserves history. Differentiates from similar actions like resolving issues.

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 instructs not to use for resolving underlying issues and to take action first. Provides clear when-not-to-use 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 are consistent (idempotentHint=true, readOnlyHint=false). Description adds critical behavioral context: the tool only applies one-click recommendations and should not be used for unvalidated changes. No 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?

Three sentences: purpose, condition, and usage guidance. No filler, front-loaded with essential 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?

Given the tool's simplicity (3 params, no output schema), the description fully covers what it does, when it works, and when to use alternatives. No missing crucial details.

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 provides examples of optimizations (max_tokens, model switch) but does not add detail beyond schema descriptions for the parameters themselves.

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 'Apply' and the resource 'optimization recommendation to an agent'. It specifies the condition 'one_click: true', distinguishing it from other tools.

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 warns 'Do NOT use for unvalidated changes' and recommends an alternative tool 'create_model_experiment' for uncertain cases. Also states that it only works for one_click suggestions.

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=false and destructiveHint=false, so the description's claim of creating a mapping is consistent but adds minimal behavioral context beyond what annotations provide.

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 wasted words. The purpose is front-loaded, and the usage guideline is concisely appended.

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 6 parameters and no output schema, the description covers purpose and usage exclusions adequately. It could mention the return value or confirmation, but the schema and annotations fill many 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 100%, so the schema already documents each parameter well. The description does not add additional meaning beyond the schema's definitions.

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 ('Link'), the resource ('agent task/event to a business outcome'), and the purpose ('for ROI tracking'). It distinguishes itself from sibling tools by explicitly mentioning 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?

The description explicitly states when not to use this tool ('Do NOT use for reading attribution data') and provides specific alternative tools ('use get_attribution_report or get_task_roi').

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, destructiveHint=false, idempotentHint=true. The description adds value by specifying that the tool returns cost savings estimates and works without usage data (Day 0). No contradictions; adds useful behavioral context beyond 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?

Two concise sentences: first states purpose and returns, second provides usage guidance with sibling alternative. 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?

Despite no output schema, the description lists returned data (pricing, context window, savings) and the use case. Combined with annotations (readOnly), it provides a complete picture for usage. All relevant signals covered.

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?

Input schema coverage is 100%, so each parameter already has a description. The description adds overall context but does not significantly enhance parameter understanding beyond the schema. 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 clearly states the tool compares LLM model pricing and capabilities across providers, listing specific return values (pricing per 1M tokens, context window sizes, etc.) and highlighting 'Day 0 value'. This is a specific verb-resource pair that distinguishes it from siblings like get_optimization_recommendations.

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 'Do NOT use for agent-specific recommendations' and directs to an alternative sibling tool (get_optimization_recommendations), providing clear when-to-use and when-not-to-use 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?

The description adds behavioral context beyond annotations: alerts can trigger email, webhook, or auto-pause, and thresholds run server-side. Annotations idempotentHint=true is consistent with configure action; no 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?

Three sentences, each adding distinct value: purpose, actions/behaviors, and exclusion. No wasted words; front-loaded with 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?

Covers purpose, usage, behavior, and parameter implications. Minor gap: no mention of return value (since no output schema), but otherwise complete for a configuration 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 baseline is 3. The description mentions actions and threshold values but largely repeats information already in parameter descriptions. It does not add significant new meaning 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 the tool sets up cost or operational alert thresholds for an agent or org-wide, using specific verbs and resource identification. It explicitly distinguishes from the sibling tool 'get_alerts'.

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 when-to-use ('Set up... thresholds') and when-not-to-use ('Do NOT use for viewing current alerts — use get_alerts instead'), plus the use case context ('real-time cost governance and operational safety').

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 key behaviors: routing traffic, tracking metrics, running until significance or max duration, and auto-promote option. Annotations provide no behavioral hints, so the description carries the full burden. It is thorough but could mention what happens if not auto-promoted or how to stop. 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 three sentences with no wasted words. It front-loads the main purpose, then explains mechanics, and ends with usage guidelines. Well-structured and concise.

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 complexity (7 params, no output schema), the description covers purpose, behavior, and sibling distinction. It lacks mention of the return value (like experiment ID) and prerequisites (e.g., agent must exist). Still, it is fairly complete for an AI agent.

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 100% description coverage for all 7 parameters. The tool description summarizes the parameters' roles but does not add significant new meaning beyond the schema. 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 tool's purpose: 'Start an A/B test comparing two LLM models for a specific agent.' It also distinguishes from a sibling tool by explicitly saying 'Do NOT use for one-off model comparisons — use compare_models for static pricing data.'

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 when-to-use and when-not-to-use guidance, including an alternative tool for one-off comparisons. It says 'Routes a percentage of traffic...' and 'Do NOT use for one-off model comparisons — use compare_models for static pricing data.'

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, destructiveHint=false, idempotentHint=true. The description adds value by detailing the report contents (per-agent breakdown, confidence scores, risk flags) and suitability for board reporting, which provides context beyond the annotations about the nature of output and use cases.

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 with two sentences: the first states the purpose and contents, the second provides explicit usage guidance with a sibling alternative. No extraneous text.

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 annotations cover safety and the schema covers parameters, the description provides sufficient context about the report's purpose and contents. However, it lacks details about the output format (e.g., JSON structure), but the listed contents partially 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 description coverage is 100%, so the baseline is 3. The description does not add additional meaning beyond the schema's parameter descriptions (e.g., period_days, include_methodology, agent_ids).

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 generates a comprehensive ROI audit report for the AI agent fleet, listing specific contents (cost/revenue breakdown, confidence scores, opportunities, risk flags). It distinguishes itself from the sibling tool get_task_roi by explicitly stating when not to use and providing the alternative.

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 states the appropriate use cases (board reporting, compliance) and provides a clear exclusion: 'Do NOT use for quick per-agent ROI checks — use get_task_roi for individual agents.' This directly guides the agent on tool selection.

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 description reinforces that this is a safe read. It adds value by listing the kinds of information returned (model, framework, etc.), but does not mention error handling or response format, which is acceptable for a simple lookup.

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: one defining the action and output fields, the second providing usage guidance. Efficient and front-loaded with essential 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?

Given the simple interface (one required param, no output schema), the description adequately describes the tool's purpose and output content. However, it does not specify the return structure (e.g., single object vs array), which would complete the picture.

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 schema describes agent_id as 'The agent UUID to look up'. The description does not add additional parameter details beyond what the schema provides, so 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 clearly states 'Get detailed information about a specific agent' and lists specific fields (model, framework, category, etc.), distinguishing it from the sibling tool get_cost_summary which is for fleet-wide overviews.

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 tells when NOT to use ('Do NOT use for fleet-wide overviews') and directs to the correct alternative ('use get_cost_summary instead'). This provides clear context for selection among siblings.

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 indicate the tool is read-only, non-destructive, and idempotent. The description adds no behavioral details beyond that, but it does not contradict the annotations. It explains what the tool returns, which is sufficient given the annotations' clarity.

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 with three short sentences: the first states the purpose, the second lists alert types, and the third provides an exclusion and alternative. No extraneous 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?

Given the simple read-only tool with three optional parameters and no output schema, the description covers the core functionality, returned content, and key constraints. It is complete for its complexity level.

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 100% description coverage for all three parameters (severity, unread_only, limit), so the tool description does not need to add parameter details. The description only reiterates filtering by severity, which is already in the schema, thus no additional value.

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 retrieves active alerts and notifications for the agent fleet, listing specific alert types (cost spikes, error rate increases, budget warnings, system health). It explicitly distinguishes from the sibling tool for configuring alert triggers, ensuring no confusion.

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 an explicit when-not-to-use scenario and directs users to an alternative tool (configure_alert_threshold) for configuring triggers. However, it does not elaborate on when to use this tool over other sibling tools for similar tasks, which would make it a 5.

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 read-only, non-destructive, idempotent behavior. The description adds transparency about the 'model' parameter being informational only due to backend behavior, and references the canonical enum source. This disclosure is valuable beyond 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 extremely concise: three sentences that front-load the purpose, include the warning about alternative tool use, and add necessary behavioral notes. No superfluous 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?

Despite lacking an output schema, the description hints at return structure (outcome counts, values, etc.). With zero required parameters and all schema descriptions covered, the contextual information is sufficient for most usage scenarios. Could mention pagination or limits, but not essential.

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?

With 100% schema description coverage, the schema already describes all three parameters. The description adds value by explaining the 'model' parameter's actual effect (informational only) and providing context about the enum source. The baseline is 3, and this added insight merits a 4.

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 retrieves an attribution report linking agent actions to business outcomes, listing specific contents (outcome counts, values, confidence scores, top agents). It also distinguishes from the sibling 'generate_roi_audit' by explicitly excluding its use for board-level reporting.

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 not to use this tool ('Do NOT use for board-level reporting') and directs to an alternative ('use generate_roi_audit'). It lacks an explicit 'when to use' statement but the context of attribution analysis 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 declare readOnlyHint=true, destructiveHint=false, and idempotentHint=true. The description adds context about what the tool shows (spending vs limits, warning/exceeded counts, enforcement modes) and confirms it accesses all budget configurations, aligning with annotations with 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 concise sentences: first states purpose and content, second provides usage guidance. No extraneous 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?

With no parameters and a straightforward read operation, the description covers what the tool returns. It is complete enough for an agent to understand the tool's function, though it does not detail the output structure (no output schema provided).

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 no need for parameter explanations. The description adds no parameter detail, but baseline 4 is appropriate per guidelines.

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 the specific verb 'Get' and resource 'budget status', clearly stating it retrieves current status of all budget configurations. It distinguishes itself from sibling tools like set_budget and update_budget_mode by explicitly noting it is read-only.

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 tells the agent exactly when to use this tool ('to monitor spending governance') and explicitly warns against using it for creating/changing budgets, directing to specific alternatives ('set_budget or update_budget_mode').

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, destructiveHint=false, idempotentHint=true. Description adds value by listing returned fields (total spend, call counts, etc.), but doesn't mention response format or pagination. 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?

Two sentences, no fluff. Purpose and usage guidance are front-loaded. 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?

Return fields are listed, usage context provided, and tool differentiated from sibling. However, no output schema exists, and the description doesn't specify whether data is cached or real-time, leaving minor 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 has 100% coverage for the single parameter 'period_days' with its description. The tool description does not add any additional meaning or nuance to the parameter, 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?

Description clearly states 'Get a comprehensive cost summary for your AI agent fleet' – specific verb and resource. Distinguishes from sibling tools by referencing real-time per-request cost checking as a separate use case.

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 tells when to use ('starting point for understanding agent economics') and when not to use ('Do NOT use for real-time per-request cost checking'), and provides alternative ('use OpenTelemetry spans').

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 agent knows it's safe. Description adds what results contain (samples, metrics, significance, winner), which is useful but not critical for safety.

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 outputs, second provides negative guidance. No wasted words, immediately 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 result tool with good annotations and full schema coverage, the description covers the key outputs and usage boundary. Lacks mention of real-time vs cached data, but overall sufficient.

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 description carries little burden. It does not elaborate on parameter meanings, but the schema descriptions are sufficient. No additional value 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 verb 'Get' and resource 'experiment results' and lists specific outputs (sample counts, metric comparisons, etc.). It also distinguishes from starting experiments, providing clear purpose.

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 states when not to use ('Do NOT use for starting experiments') and provides the alternative tool ('use create_model_experiment'). This is excellent 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 already declare readOnlyHint=true, destructiveHint=false, idempotentHint=true, covering safety. Description adds context about output content (predictions with confidence and recommendations), which is not in schema. No contradictions, but could mention pagination or rate limits; still strong.

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 wasted words. Front-loaded with purpose, followed by usage guidance. Structure is clear and efficient.

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 query tool with three optional parameters and no output schema, the description adequately explains what the tool returns (predictions with confidence and recommendations) and distinguishes from related tools. No missing critical information.

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 schema already fully describes the three optional parameters (agent_id, severity, status). Description does not add parameter details beyond the schema, meeting the baseline for high coverage.

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 'Get' and resource 'predictive failure analysis for your agents', listing specific risk types (error rate breaches, latency degradation, etc.). It explicitly distinguishes from siblings by advising not to use for current/past failures and referencing get_alerts.

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 on when to use (upcoming risks) and when not to (current/past failures), with an alternative tool mentioned (get_alerts). Also notes that predictions include confidence level and recommended actions, aiding decision-making.

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 idempotentHint. Description adds context about output content (estimated monthly savings, confidence level) and reinforces read-only nature, but 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?

Three sentences, no fluff. Front-loaded purpose, lists content, then provides usage boundaries. Efficient and well-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?

Describes output partially (suggestions with savings and confidence) and includes usage boundaries. Without output schema, this is adequate; minor gap in not stating exact structure.

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 description adds no significant parameter info beyond existing schema descriptions. 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?

Description clearly states 'Get AI-powered cost optimization recommendations' with specific examples (model switching, provider arbitrage, etc.) and distinguishes from sibling tools like apply_optimization.

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 'Do NOT use for implementing fixes' and directs to apply_optimization or create_model_experiment, providing clear when-to-use and when-not-to-use 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 already declare readOnlyHint=true, destructiveHint=false, idempotentHint=true. The description adds value by detailing the output structure (costs, outcomes, breakdown), but does not reveal additional behavioral traits beyond what annotations imply. No 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: first declares purpose, second lists output and exclusion. Front-loaded, no extraneous words, 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 tool's simplicity, rich annotations, and no output schema, the description covers purpose, usage guidelines, and high-level output. Slightly incomplete on return format, but sufficient for an experienced user.

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 has 100% description coverage, so baseline is 3. The description does not add extra meaning to parameters beyond the schema; it focuses on output rather than parameter details.

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 calculates ROI for an agent, listing specific metrics (costs, outcomes, multiplier, breakdown) and explicitly distinguishes from the sibling tool generate_roi_audit for fleet-wide analysis.

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 when-to-use ('calculating ROI for an agent') and when-not-to-use ('Do NOT use for fleet-wide ROI'), with a direct sibling reference. Also notes usefulness for identifying high-value agents.

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 tool as read-only, non-destructive, and idempotent. The description adds behavioral context: it analyzes usage patterns, calculates optimization potential, and provides a structured recommendation with projected savings. This goes beyond the annotations and informs the agent about the tool's output and processing.

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: the first covers purpose and functionality, the second gives a clear negative use case. No unnecessary words, well-structured with key information upfront.

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 tool with one optional parameter and no output schema, the description covers purpose, process, and a usage guideline. It distinguishes from related siblings. A minor gap is the lack of detail on the output format, but it is still 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 description coverage is 100% for the only parameter (period_days). The description does not add any new semantics beyond what the schema already provides. According to the rubric, baseline is 3.

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 generates an ROI report for upgrading from Starter to Lite/Pro tier. It uses a specific verb ('Generate') and resource ('ROI report'), and the context of 'upgrade justification' distinguishes it from sibling tools like metrx_generate_roi_audit.

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 'Do NOT use if already on Lite or Pro tier — not relevant for paid-tier users.' This provides clear exclusion criteria. However, it does not name specific alternative tools for paid-tier users, which would elevate it to a 5.

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 destructiveHint=false, so the description does not need to repeat safety. However, it adds valuable context that the tool 'Returns agent IDs needed for other tools', which informs the agent about how the output may be used downstream.

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 unnecessary words. The key information (what the tool does and when to avoid it) is front-loaded. Every sentence serves a 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?

For a simple list tool with two optional parameters and no output schema, the description covers everything needed: what is returned, that IDs can be used elsewhere, and the boundary with the detail tool. Annotations handle safety and idempotency.

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% (both parameters have descriptions). The description merely reiterates that filtering is optional ('Optionally filter by status or category') without adding new meaning 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 uses a specific verb ('List') and resource ('AI agents'), and clarifies what data is returned (status, category, cost). It also distinguishes from the sibling tool 'get_agent_detail' by stating that this is for listing, not detailed analysis.

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 states when not to use this tool ('Do NOT use for detailed per-agent analysis') and provides a clear alternative ('use get_agent_detail for that'). This helps the agent select the correct 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 already declare readOnlyHint=true and destructiveHint=false. The description adds behavioral context: it uses historical performance data and cost analysis, and suggests optimal model per complexity level. This complements the annotations well.

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 effectively convey purpose, mechanism, and usage boundaries. No filler; 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?

While the description covers purpose and usage well, it lacks details about the tool's output (return value format). With no output schema, specifying what the recommendation looks like would be helpful.

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 baseline is 3. The description does not add new parameter-level details beyond what the schema provides, though it aligns with the task_complexity parameter.

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 gets a model routing recommendation for a specific task based on complexity, using historical data and cost analysis. It distinguishes itself from the sibling tool compare_models by explicitly excluding that use case.

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 this tool ('for a specific task based on complexity') and when not to ('Do NOT use for comparing all models at once'), directly naming the alternative tool (compare_models).

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, destructiveHint, and idempotentHint. The description adds context about the 7 inefficiency types and output format options, but the behavioral safety is already well covered by annotations. The description does not contradict 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, well-structured paragraph. It front-loads the main purpose, then lists inefficiencies, output description, and usage caveats. Every sentence adds value with no repetition.

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 no output schema, the description sufficiently describes the return value (scored report with fix recommendations and estimated savings). It also explains the output format options and the scope of the audit. The description is complete for an agent to understand tool behavior.

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 each parameter has a clear description. The description adds value by explaining the purpose of output_format for CI/CD pipelines and clarifying that agent_id is for scanning a specific agent. This goes beyond the 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 runs a comprehensive cost leak audit across the agent fleet, identifies 7 specific inefficiencies, and returns a scored report with recommendations. It distinguishes itself from sibling tools like configure_alert_threshold and apply_optimization by explicitly stating what it does and does not do.

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: 'Do NOT use as a continuous monitoring loop — use configure_alert_threshold' and 'Do NOT use for fixing leaks — use apply_optimization'. This clearly tells the agent when to use this tool and when to use alternatives.

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 explains the enforcement modes and the create/update behavior, adding context beyond annotations (readOnlyHint=false, idempotentHint=true). It does not contradict annotations. Minor gap: it does not explicitly state whether calling it overwrites all existing budget 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?

The description is two sentences with no wasted words. The core purpose is front-loaded, and every sentence serves a clear purpose (purpose + usage guidance).

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 no output schema, the description does not mention return values or side effects. It adequately covers input parameters and usage restrictions, but the lack of response information is a gap for a tool that modifies state.

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%. The description adds value by explaining enforcement modes with examples, clarifying the agent_id omission for org-wide budgets, and giving a dollar format example, all 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 the tool creates or updates a budget configuration for an agent or organization, using specific verbs and resource. It also distinguishes itself from the sibling 'update_budget_mode' by explicitly warning against using it solely for mode changes.

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 not to use the tool (just to change enforcement mode) and points to the correct alternative (update_budget_mode). However, it does not discuss when to use this tool over other budget-related siblings like get_budget_status.

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 (idempotentHint=true, destructiveHint=false) align with description. Description adds that results are preserved and optional promotion, providing full behavioral disclosure beyond 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 three sentences, front-loaded with the action, and every sentence adds value without redundancy.

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 tool's simplicity (2 params, no output schema), the description fully covers purpose, usage, behavior, and parameter semantics. No gaps remain.

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 context to promote_winner ('If the treatment model won, you can optionally promote it'), clarifying its purpose beyond the schema default 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 clearly states 'Stop a running model routing experiment' with a specific verb and resource. It distinguishes from pausing and mentions results preservation, differentiating from sibling tools like pausing 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?

Explicitly says 'Do NOT use for pausing experiments temporarily — stopping is permanent,' providing clear when-not-to-use guidance and an alternative 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 indicate non-read-only, non-destructive, idempotent behavior. The description adds context about pausing/resuming and mode meanings but does not disclose potential side effects or error handling. It is consistent 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, then mode guidance, then exclusion. No superfluous words; each 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?

Covers the core functionality and usage boundaries. For a tool with three parameters and no output schema, it is largely complete. Could mention what happens on invalid budget_id, but this is acceptable.

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 three enum modes ('alert_only' for monitoring, etc.), which goes beyond the schema's minimal 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 the verb (change/pause/resume) and the resource (budget), and distinguishes from sibling tool 'set_budget' for creation. It specifies the exact actions and modes.

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 tells when to use each mode ('alert_only for monitoring', etc.) and includes a clear exclusion: 'Do NOT use to create new budgets — use set_budget for that.' This provides strong guidance for the agent.

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