lookup_local_modelCherche un modèle dans le catalogue public OutilsIA (nom, taille ou tag Ollama). Ce n'est pas un scan et l'outil n'invente aucun tokens/s.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes |
| Name | Required | Description | Default |
|---|---|---|---|
| lookup | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations already declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, establishing a safe, bounded read operation. The description adds one detail: it does not invent tokens/s (meaning no hallucination), which is a useful behavioral note beyond annotations. However, it omits other relevant behaviors such as whether it accesses a live API or local cache, rate limits, or result format. With annotations covering safety, a 3 is appropriate.
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 zero wasted words. Every sentence adds necessary context: the first covers the core purpose and searchable dimensions, the second clarifies it's not a scan and doesn't generate fabricated data. This is exemplary conciseness.
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?
The tool has one simple string parameter, high-coverage annotations, an output schema, and is part of a sibling group focused on hardware/report tools. The description adequately explains the search action and the public catalog scope. However, it doesn't describe what the output contains (though an output schema exists, reducing the burden), and it doesn't address data freshness or query constraints beyond maxLength. For a single-param lookup, this is minimally adequate but not thorough.
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 a single parameter (query) with 0% schema description coverage, meaning the JSON schema provides no semantic hints beyond type and length constraints. The description partially compensates by listing search dimensions (name, size, or Ollama tag), which gives the agent guidance on what 'query' can contain. However, this is minimal—it doesn't explain formatting, case sensitivity, or how multiple dimensions combine. Baseline 3 is reasonable given the low coverage but some added 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 verb 'Cherche' (search) and the resource 'dans le catalogue public OutilsIA' with specific search dimensions (name, size, or Ollama tag). It also explicitly distinguishes itself from siblings by stating 'Ce n'est pas un scan', which is unique among the sibling tool names that revolve around reports, benchmarks, and hardware 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?
The description states what the tool does and that it searches a specific public catalog, providing basic context. However, it does not specify when to prefer this over a sibling tool (e.g., when checking for model availability vs. analyzing a runtime report), nor does it mention any conditions for non-use or alternatives. The 'no invention of tokens/s' is a negative clarification but not a usage directive.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool has a distinct purpose (analyze, check, explain, list, etc.) and the detailed descriptions make boundaries clear. However, the difference between 'reading a shared report' (analyze_shared_report) and 'reading a shared report to list installed models' (list_installed_models_from_report) or 'reading a shared report to list benchmarks' (list_benchmark_proofs_from_report) could cause an agent to pick the wrong one. There is also some overlap between these list operations that read a report versus simply displaying a pre-generated cockpit (render_machine_cockpit).
The majority of tools follow a consistent verb_noun pattern (e.g., analyze_shared_report, explain_bottleneck, list_installed_models_from_report, recommend_runtime). Minor inconsistency exists with the use of 'geo_audit' vs 'geo_kit' vs 'ratings' and the verb tense in 'list_benchmark_proofs_from_report' and 'list_first_party_measurements' departs from a simple pattern.
15 tools is at the high end of the ideal range (3-15), but each tool appears justified given the comprehensive scope of local AI assistance (hardware checking, model lookup, benchmarking, reporting, educational/reference tools). Slightly over-stuffed but still manageable for an agent.
The surface covers a complete workflow: check hardware, lookup models, benchmark, generate cockpit, explain bottlenecks, simulate upgrades, and reference documentation. The missing piece is the lack of 'update' or 'delete' operations, but this is by design, as the entire workflow is read-only. The set seems like a complete view of all possible read-only interactions with the domain.