Server Quality Checklist

Tool Scores

• ollama_lifecycleA
  4.2/5.0
  Behavior4/5

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

  Discloses meaningful behavioral context beyond annotations: pre-loading skips ~12s cold-load cost, unload frees VRAM immediately, and large models can consume tens of GB. Consistent with annotations (idempotentHint true, readOnlyHint false, destructiveHint false). 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.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Three concise sentences, front-loaded with the overall purpose. Each sentence adds distinct value: action breakdown, warm use case, unload resource impact. No filler or redundancy.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Covers the three action variants, their resource implications, and return behavior for status (VRAM use, time until unload). Lacks explicit return descriptions for warm/unload, but no output schema exists and the side effects are clearly explained. Sufficient for a management tool with good annotations.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Input schema has 100% parameter coverage, so the schema already explains each parameter in detail. The description adds some context (e.g., warm/unload need `model`, keep_alive duration impact) but mostly relies on the schema. Baseline 3 is appropriate.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description states the tool's core function ('Manage model residency') and details each action with specific verbs: `status` lists loaded models, `warm` pre-loads, `unload` frees VRAM. It clearly distinguishes itself from siblings like dispatch (inference) and models (management) by focusing on runtime residency in VRAM.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides concrete use cases: warm is 'useful before a batch', unload matters because a large model 'can hold tens of GB and starve everything else'. It implies using status to inspect residency, but does not explicitly name alternative tools or state when not to use this tool. Clear context, but lacks explicit exclusions.

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

• ollama_dispatch_batchA
  4.6/5.0
  Behavior5/5

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

  Beyond annotations, the description reveals non-obvious execution semantics: grouping by resolved model, sequential group execution to avoid VRAM thrashing, input-order result return, per-item failure isolation, and context cost savings. These traits are not encoded in the annotations (readOnlyHint, openWorldHint, etc.) and materially shape the agent's expectations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Four concise sentences, each adding distinct value: the core purpose, grouping/VRAM rationale, result ordering/failure isolation, and the explicit comparison with the sibling. No redundant or filler content; every sentence earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Despite having 10 parameters and nested objects, the description covers the most important behavioral context—batching rationale, ordering, failure handling, and context cost—while the schema documents parameter details in depth. It does not explicitly describe the result item structure, but the item schema's 'id echoed back' and 'results returned in input order' imply a list of per-item outputs, so the gap is minor.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 100% with rich descriptions for all 10 parameters (e.g., model grouping, concurrency, keep_alive). The description's high-level statements about grouping and ordering are already replicated in the schema's parameter descriptions, so it adds no new parameter-level meaning; baseline 3 applies.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description opens with 'Run many generations in one call', a specific verb+resource statement. It clearly distinguishes from sibling 'ollama_dispatch' by explicitly recommending batching over multiple separate calls, establishing its scope as the batch variant.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines5/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Explicitly instructs to 'Prefer this over many separate ollama_dispatch calls' when handling many prompts, naming the alternative and the concrete benefits (faster and less context). This gives clear when-to-use guidance and an implicit when-not-to-use for single dispatch, matching the level of the calibration example for get_calls.

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

• ollama_dispatchA
  4.7/5.0
  Behavior5/5

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

  The description goes far beyond the basic annotations, disclosing that reasoning and answer tokens share one num_predict budget, thinking is off by default, files are read server-side and never enter the caller's context, resolution never silently substitutes unknown models, and output trimming is reported. This is substantial behavioral disclosure with no contradiction against annotations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is dense and front-loaded, with the first sentence stating purpose immediately. Every subsequent sentence adds unique information—use cases, context protection, model resolution, token budget, and output metrics—with no filler or repetition.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness5/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the tool's complexity (26 params, no output schema) and sparse annotations, the description is remarkably complete: it covers purpose, when to use, model resolution, file handling, security constraints, token-budget caveats, and what the response includes. The schema handles individual parameter details, and the description fills the behavioral and selection gaps.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 100% schema description coverage, the baseline is 3, but the description adds valuable context beyond the schema: the three model-selector grammars, the token-saving rationale behind files/file_globs, and the warning about num_predict sharing with thinking tokens. This elevates it above the baseline without being redundant.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The opening sentence is a specific verb+resource statement: 'Run one generation on a local Ollama model and return its output.' It clearly differentiates from siblings by emphasizing a single generation, and the rest of the description enumerates concrete use cases like summarization and extraction.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description gives clear context for when to use it ('mechanical work you do not want to spend your own context on') and highlights the file-reading feature. However, it does not explicitly name alternatives or state when NOT to use it, such as pointing to ollama_dispatch_batch for multiple generations.

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

• ollama_modelsA
  4.7/5.0
  Behavior5/5

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

  Beyond the readOnlyHint and idempotentHint annotations, the description reveals behavioral details: explain_selector 'without spending a generation', the reporting of the 'full fallback chain', and the need to refresh after pulling a model. This adds real context about system behavior and costs that annotations alone do not convey.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is three sentences, each serving a distinct purpose: the first states the core list function and output, the second covers filtering and refresh, and the third explains explain_selector with a usage recommendation. There is no redundant or filler content.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness5/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The description covers the primary list function, filtering, refresh, and the advanced explain_selector feature, and also describes the output fields (capabilities, context window, size, loaded status). Since there is no output schema, this is sufficient. The tool's complexity is handled well because the schema documents all parameters and the description focuses on when and how to use them.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 100%, so the baseline is 3. The main description adds operational meaning: it links `refresh` to post-pull actions, connects `resident_only` to avoiding load costs, and frames `explain_selector` as the recommended debugging entry point with the 'full fallback chain' detail. This is more than just restating the schema, though not dramatically so.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description opens with a specific verb and resource: 'List the models installed on the Ollama host' and enumerates the output fields (capabilities, context window, size, loaded status). It also clearly introduces the second function, explain_selector, as a dry-run of the model resolver, which distinguishes it from siblings like ollama_dispatch and ollama_lifecycle.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description provides clear conditional usage: 'pass `refresh` after pulling a new model', 'Filter with `capability` or `resident_only`', and for explain_selector, 'Start there whenever routing surprises you.' It gives context and a recommendation, but does not explicitly name alternative tools or state when not to use the tool.

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