PV Solaire Énergie — Estimation solaire et devis (France)

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

Beyond the annotations, the description discloses the 30-minute code validity, the five-attempt limit, and the fact that this validation is the sole authorization for forwarding the request to installers. It also explains the no-duplicate consequence of retrying demander_devis, adding valuable behavioral context.

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 dense sentences cover purpose, workflow, user interaction, and edge-case handling without redundancy. The most important information is front-loaded.

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 tool with no output schema, the description supplies all necessary workflow context: input provenance, retry conditions, expiration, and downstream side effect. Nothing needed to invoke it correctly is missing.

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 schema already covers both parameters at 100%, so the baseline is 3; the description adds meaningful provenance semantics by stating demande_id comes from demander_devis, that the code must be requested from the user rather than guessed, and that retries do not create duplicates.

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 ('Valide une demande de devis') with a precise mechanism (4-digit SMS code) and names the related tool that creates the request. It clearly distinguishes itself from demander_devis, and the title reinforces the same 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?

It explicitly instructs the agent to ask the user for the code, call the tool with the demande_id from demander_devis, and provides the exact fallback flow if the code expires or after five failed attempts. This is model guidance on when and how 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.

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

Annotations only declare the generic write/open-world profile; the description adds the concrete side effects: an SMS 4-digit code is sent, a `demande_id` is returned, nothing reaches an installer before confirmation, and the consent/real-data requirement. That is exactly the kind of behavior an agent cannot infer from readOnlyHint=false.

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?

Front-loaded with purpose, then the hard warning, then the workflow and prerequisites. Every sentence is actionable (state the code flow, never claim it was sent, consent rules, call estimer first); nothing is filler.

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 17-parameter write tool with no output schema, the description still covers the return value (`demande_id`), the failure mode (fixed or wrong number permanently blocks the request), the prerequisite tool, and the confirmation tool. An agent has everything needed to call it and to explain the outcome to the 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 coverage is 100%, so the baseline is 3, but the description adds cross-parameter meaning the schema does not: `estimation_id` is strongly recommended and valid for 2 hours, which makes `orientation` and `surface_toit` redundant. It does not, however, give per-field guidance for the remaining 15 parameters.

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?

States a specific verb and resource ('Dépose une demande de devis gratuite... pour une installation de panneaux solaires en France') and immediately scopes it as step one of a two-step flow, naming `confirmer_demande` as the required follow-up. An agent can distinguish it from `confirmer_demande`, `estimer_production_solaire` and `rechercher_installateurs` without opening any schema.

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?

Explicit when-to-use (with explicit consent, real contact data, no test data), explicit ordering constraint (the request is not transmitted to any installer until `confirmer_demande` validates the SMS code), and an explicit recommendation to call `estimer_production_solaire` first. Also tells the agent what not to announce to the user.

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 openWorldHint. The description adds meaningful behavioral context: the analysis is based on real 3D geometry (Google Solar), computed via PVGIS, and explicitly states that results are not a regional average or a rule-of-three. It also discloses that the heat map is returned only 'quand elle est disponible', consistent with openWorldHint.

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 focused paragraph with a strong front-loaded action and output list. The clarification 'ce n'est ni une moyenne régionale ni une règle de trois au m²' earns its place by setting expectations. It is slightly long but every sentence adds value; a 4 reflects good structure with minor room for tighter phrasing.

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 description covers the input condition (French postal address), the computation method, the full set of returned fields, the optional heat map, the estimation_id link to demander_devis, and explicit usage triggers. Since there is no output schema and no other structured description of results, this text carries that burden and does so thoroughly for a 2-parameter read-only 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 schema descriptions are already rich, including an example for adresse and a note that type_batiment changes results. The main description emphasizes 'à partir de sa seule adresse postale' but adds no new meaning about either parameter 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 opens with a specific verb ('Calcule') and a precise resource ('le potentiel photovoltaïque réel d'une toiture'), then lists concrete output fields (kWc, kWh, panneaux, orientation, etc.). It clearly distinguishes itself from siblings by mentioning the estimation_id passed to demander_devis, leaving little ambiguity about its role.

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 to use the tool: 'Utilise cet outil dès qu'un utilisateur demande combien produirait une installation solaire chez lui...' It also gives contextual guidance in the schema about requesting a precise address before calling. However, it does not explicitly say when not to use it or name alternatives, so it stops short of 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 only declare readOnlyHint and openWorldHint; the description adds real behavioral context: results are restricted to professionals whose RGE qualification is still valid today, ordered by increasing distance, and it lists the returned fields (name, city, distance in km, qualification end date, profile URL). It even explains the regulatory stakes of the RGE filter (EDF OA surplus sale, 5.5% VAT). No pagination or rate-limit note, and no output schema exists to cover the return shape, but the description carries it 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 front-loaded sentences that go from what is searched, to the RGE filter and return fields, to when to use it. The regulatory parenthetical about TVA 5.5% and the end of the autoconsumption prime is tangential but arguably supports agent reasoning about RGE relevance; it is the only expendable material.

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?

No output schema exists, yet the description enumerates the returned fields, the filtering rule, the ordering and the data source, which is everything an agent needs to call it and interpret results. The tool is a read-only search with full parameter documentation, so nothing material is missing.

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 already carries an inline description with examples, enum meanings and defaults. The description only restates the 'commune or postal code' input and adds the default sort, so it does not meaningfully exceed what the schema provides. Baseline 3 is correct.

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?

States a specific verb and resource (search a directory of photovoltaic installers and engineering firms around a commune/postal code), names the source directory, the sort order (distance ascending) and the RGE-validity filter. An agent can distinguish it from demander_devis or estimer_production_solaire without opening the schema.

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 lists three triggering situations: finding a local solar professional, checking whether certified installers exist nearby, or comparing companies before asking for a quote. The phrase 'avant de demander un devis' implicitly points to the demander_devis handoff, but no sibling is named as an exclusion, so it stops short of full routing guidance.

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