explain_bottleneckIdentifie le goulot probable à partir d'un rapport /r/... ou d'un profil RAM/VRAM déclaré. Ce n'est pas un scan. Fournir report_url ou bien ram_gb et vram_gb.
| Name | Required | Description | Default |
|---|---|---|---|
| ram_gb | No | ||
| vram_gb | No | ||
| cpu_name | No | ||
| gpu_name | No | ||
| cpu_cores | No | ||
| report_url | No | ||
| unified_memory | No | ||
| storage_free_gb | No |
| Name | Required | Description | Default |
|---|---|---|---|
| bottleneck | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and destructiveHint=false, so the description's job is lighter. The description adds value by stating 'not a scan' (behavioral mode) and describing the input method. No contradictions with annotations. Some behavioral context (e.g., what happens with incomplete data) is missing but acceptable given the 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 short sentences, front-loaded with the core action and input instruction. Every word earns its place; no redundancy or filler. Ideal conciseness for a tool with a straightforward 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?
Given 8 parameters (none required, many optional) and an output schema, the description only covers the minimal input scenario. It does not explain the role of the other parameters (cpu_name, gpu_name, etc.) or how they interact. The output schema exists, so return values need not be described, but the input semantics are incomplete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It mentions only 3 of 8 parameters (report_url, ram_gb, vram_gb) and gives partial semantics (alternative inputs). The other 5 parameters are unmentioned, leaving the agent to infer their role or ignore them. This is insufficient compensation for the low 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 tool identifies a bottleneck from a report or RAM/VRAM profile, using a specific verb ('identifies') and resource ('bottleneck'), and clarifies it's not a scan. However, it does not explicitly distinguish itself from sibling tools like 'analyze_shared_report' or 'check_pc_for_local_ai', which could overlap in 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?
The description instructs to provide either 'report_url' or 'ram_gb' and 'vram_gb', which gives clear entry conditions. The note 'not a scan' implies it should not be used for scanning. However, no explicit when-not-to-use or alternatives to sibling tools are provided.
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.