Skip to main content
Glama

list_decisions

Read-only

Returns the project's decision REGISTER in ONE call: what was decided, by whom, when, with what verbatim — and what is IN FORCE (inForce). Useful before proposing or re-debating a topic: a decision in force is not reopened without saying so. Two sources, PROJECTED (nothing is stored separately, no migration): (1) ARBITRATION tasks (an arbitration is a question with options posed to a human) — maturity derived from the step of the cycle (propose/conteste → a_trancher · arbitre → tranchee · en_cours → en_cours · clos → actee · caduc → caduque), with origin "panel" (approved in the panel) or "external" (decision made outside the panel and recorded with proof by record_external_decision: verbatim, source, real date); (2) ORDINARY tasks that carry the decision label (case and accents ignored), origin "task" — their stage comes from a stade:idee|cadrage|a_trancher|tranchee|en_cours|actee label, otherwise from the status (todo → a_trancher, doing → en_cours, done → actee; archived without a stage → caduque). inForce = tranchee, en_cours or actee. ⚠️ NO FALSE "IN FORCE": a badly set label (unknown stage, contradictory stages, blocked task) returns a_trancher and is COUNTED in unclassified with its reason. A CLOSED arbitration task whose cycle never recorded the answer is NOT a decision: it is listed separately in unrecorded (the decision may have been made elsewhere; the body, documents or comments may carry it). counts.byMaturity gives the full distribution (before the maturity filter). Sort: decision date descending, undated last. ⚠️ WHAT IT DOES NOT DO, stated in the response's notIncluded: the label convention IS a convention (fragile by nature); it does not see a decision made in a document, a comment or the chat without a task or arbitration; nor the tasks of linked projects; it returns only a truncated summary of a task body (long reasoning is read with list_tasks fields:['body']); it neither imports nor creates any decision — to record one: create_tasks with arbitrage, record_external_decision, or a task labeled decision. decidedBy is the identifier of the account that decided in the panel (it is never a name). Modifies nothing.

Input Schema

NameRequiredDescriptionDefault
limitNoNumber of decisions returned (default: 50, max: 200).
queryNoSearch by WORDS: each word (at least 2 characters, separated by spaces; the words are never matched as an exact phrase) must appear AT THE START OF A WORD in one of the fields; case and accents ignored, "revalider" finds "re-valider". Fields read: title, labels, body, question, option labels, recommendation justification, decision note, challenge, clarification, closure summary, verbatim of an out-of-cycle decision; account and key identifiers are never searched. Without a word of at least 2 characters: refused.
cellIdNoLimits to the decisions carried by THIS block AND its child blocks (this block is never returned alone). Omitted: the whole project.
offsetNoPagination — number of decisions to skip (default: 0).
inForceNo`true`: only the decisions IN FORCE (maturity tranchee, en_cours or actee — a human decided and it is not caduque). `false`: the COMPLEMENT (idee, cadrage, a_trancher, caduque). Omitted: all.
maturityNoOne maturity or a list: idee · cadrage · a_trancher · tranchee · en_cours · actee · caduque. Combines with `inForce` (both must hold).
expectedProjectIdNoUUID of the project targeted by THIS call. ACTUALLY targets this project instead of the session project. Refused without changing anything if access is not granted.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changed
    • changedInput schema / properties / cellId / description
      Previous value: -"Borne aux décisions portées par CE bloc ET ses blocs enfants (jamais ce bloc seul). Omis : tout le projet."New value: +"Limits to the decisions carried by THIS block AND its child blocks (this block is never returned alone). Omitted: the whole project."
    • changedInput schema / properties / expectedProjectId / description
      Previous value: -"UUID du projet visé par CET appel. Cible RÉELLEMENT ce projet au lieu du projet de session. Refusé sans rien modifier si l'accès n'est pas accordé."New value: +"UUID of the project targeted by THIS call. ACTUALLY targets this project instead of the session project. Refused without changing anything if access is not granted."
    • changedInput schema / properties / inForce / description
      Previous value: -"`true` : seulement les décisions EN VIGUEUR (maturité tranchee, en_cours ou actee — un humain a tranché et elle n’est pas caduque). `false` : le COMPLÉMENT (idee, cadrage, a_trancher, caduque). Omis : toutes."New value: +"`true`: only the decisions IN FORCE (maturity tranchee, en_cours or actee — a human decided and it is not caduque). `false`: the COMPLEMENT (idee, cadrage, a_trancher, caduque). Omitted: all."
    • changedInput schema / properties / limit / description
      Previous value: -"Nombre de décisions rendues (défaut : 50, max : 200)."New value: +"Number of decisions returned (default: 50, max: 200)."
    • changedInput schema / properties / maturity / description
      Previous value: -"Une maturité ou une liste : idee · cadrage · a_trancher · tranchee · en_cours · actee · caduque. Se combine avec `inForce` (les deux doivent être vrais)."New value: +"One maturity or a list: idee · cadrage · a_trancher · tranchee · en_cours · actee · caduque. Combines with `inForce` (both must hold)."
    • changedInput schema / properties / offset / description
      Previous value: -"Pagination — nombre de décisions à sauter (défaut : 0)."New value: +"Pagination — number of decisions to skip (default: 0)."
    • changedInput schema / properties / query / description
      Previous value: -"Recherche par MOTS : chaque mot (2 caractères au moins, séparés par des espaces — jamais une phrase exacte) doit figurer EN DÉBUT DE MOT dans l’un des champs ; casse et accents indifférents, « revalider » trouve « re-valider ». Champs lus : titre, étiquettes, corps, question, libellés des options, justification de la recommandation, note de décision, contestation, clarification, résumé de clôture, verbatim d’une décision hors cycle — jamais un identifiant de compte ni de clé. Sans mot d’au moins 2 caractères : refusée."New value: +"Search by WORDS: each word (at least 2 characters, separated by spaces; the words are never matched as an exact phrase) must appear AT THE START OF A WORD in one of the fields; case and accents ignored, \"revalider\" finds \"re-valider\". Fields read: title, labels, body, question, option labels, recommendation justification, decision note, challenge, clarification, closure summary, verbatim of an out-of-cycle decision; account and key identifiers are never searched. Without a word of at least 2 characters: refused."
  2. First observed

TDQS

A4.7/5.0
Behavior5/5

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description reinforces this with 'Modifies nothing' while adding substantial behavior beyond annotations: the projection model (nothing stored separately, no migration), the maturity derivation rules, the deliberate no-false-in-force safeguard counting bad labels into `unclassified`, the `unrecorded` bucket for closed arbitrations, and sort order.

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

Conciseness3/5

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

Purpose is front-loaded in the first clause, but the body is a dense wall of text with heavy capitalization and multiple embedded warnings. The complexity justifies length, yet the same facts (label fragility, projection, exclusions) recur and could be tightened, so it reads as adequate rather than disciplined.

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?

For a highly complex tool with no output schema, the description is complete: it describes the return shape (`counts.byMaturity`, `unclassified` with reasons, `unrecorded`, truncated task bodies), the response-level `notIncluded` list, and the `decidedBy` identifier caveat — everything an agent needs to call and interpret it correctly.

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, and the schema already documents limit/query/cellId/offset/inForce/maturity. The description earns above baseline by explaining how `maturity` values are derived from the arbitration cycle steps and the `stade` label fallback, which clarifies what each enum value actually means to the caller.

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?

States a specific verb and resource — 'Returns the project's decision REGISTER in ONE call' — and immediately defines what that register contains (what was decided, by whom, when, inForce). It also names the sibling tools it is not (list_tasks for bodies, create_tasks/record_external_decision for recording), so an agent can distinguish it without opening another schema.

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?

Explicit when-to-use ('useful before proposing or re-debating a topic: a decision in force is not reopened without saying so') plus explicit when-not and alternatives: recording is routed to create_tasks with `arbitrage`, `record_external_decision`, or a `decision`-labeled task, and long reasoning to `list_tasks` `fields:['body']`. The `notIncluded` disclosure enumerates what the tool cannot see.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.