Skip to main content
Glama

Faire une action

dandori_action
Destructive

Faire une action précise sur un élément, qui n'est ni une création ni une modification de champs. Les valeurs vont dans fields ; parent_id est obligatoire quand l'action le dit. Actions :

  • complete_milestone : terminer un jalon (fields.is_completed false le rouvre)

  • advance_key_result : avancer un résultat clé de son pas (fields.direction -1 pour reculer), plan Plus requis

  • reactivate_subscription : réactiver un abonnement résilié

  • subscription_used : noter qu'un abonnement a servi (fields.used_at, par défaut aujourd'hui)

  • start_unplug_challenge : démarrer un défi de déconnexion (id du défi, voir challenges dans dandori_unplug_dashboard ; un seul à la fois)

  • abandon_unplug_challenge : abandonner le défi de déconnexion en cours (id de active_challenge dans dandori_unplug_dashboard), sans retour possible

  • merge_tags : fusionner le tag id dans fields.target_id : ses éléments passent sur la cible, puis il disparaît

  • favorite_memory : garder un souvenir, ou le retirer s'il l'était (fields : source note, task, reading ou photo, et id de l'élément)

  • remind_witness : relancer par e-mail un témoin qui n'a pas encore accepté (plan Plus requis, une fois toutes les 48 heures)

  • undo_challenge_contribution : retirer ta dernière contribution du jour à un défi commun (id du défi)

  • mark_learning_volumes : manga ou BD (id de la série) : marquer des tomes possédés ou lus, ou les retirer (« j'ai acheté le tome 13 », « j'ai les tomes 1 à 12 »), ou noter le tome lu en ce moment (« je lis le tome 12 »). fields : from (numéro du premier tome), to (du dernier, absent : le seul from), owned et read (true, false, ou absent : inchangé), current (true : from est le tome en cours, sans to ; false : rend sa place au premier tome pas lu). Un tome au delà des parus en ajoute un paru. Les tomes déjà notés : fetch learning:ID (volumes_owned, volumes_read, volume_current)

  • duplicate_workout_template : dupliquer un modèle de séance, plan Plus requis

  • link_items : lier deux éléments, dans les deux sens (fields.from_type, from_id, to_type, to_id ; types task, note, project, goal, contract, list, routine) : la page Notes filtre ensuite sur le projet, l'objectif ou la tâche. Les liens d'un élément : fetch ; en défaire un : dandori_delete, entity link et son link_id

  • send_friend_request : envoyer une demande d'ami (fields.user_id, l'ID d'un membre d'une communauté commune, voir dandori_get_community ; il faut d'abord un pseudo, choisi dans l'app)

  • set_budget_limit : fixer le plafond d'une catégorie pour un mois (fields : budget_category_id, month YYYY-MM, amount)

  • set_focus_goal : fixer l'objectif quotidien du chrono (fields.minutes, 30 à 720)

  • set_screen_time_goal : fixer la limite quotidienne de temps d'écran (fields.daily_limit_minutes, 15 au moins)

  • skip_task_occurrence : passer cette fois une tâche récurrente : elle va à sa prochaine date (next_occurrence de la tâche) sans compter comme faite ; elle compte quand même dans une fin « après N fois »

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "dandori_action"
  }
}

Input Schema

NameRequiredDescriptionDefault
idNoID de l'élément visé
actionYesL'action
fieldsNoValeurs de l'action (voir la liste)
parent_idNocomplete_milestone : ID du projet ; advance_key_result : ID de l'objectif ; remind_witness : ID du contrat ; undo_challenge_contribution : ID de la communauté

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / action / enum
      Previous value: -[
      -  "complete_milestone",
      -  "advance_key_result",
      -  "reactivate_subscription",
      -  "subscription_used",
      -  "start_unplug_challenge",
      -  "abandon_unplug_challenge",
      -  "merge_tags",
      -  "favorite_memory",
      -  "remind_witness",
      -  "undo_challenge_contribution",
      -  "mark_learning_volumes",
      -  "duplicate_workout_template",
      -  "link_items",
      -  "send_friend_request",
      -  "set_budget_limit",
      -  "set_focus_goal",
      -  "set_screen_time_goal"
      -]New value: +[
      +  "complete_milestone",
      +  "advance_key_result",
      +  "reactivate_subscription",
      +  "subscription_used",
      +  "start_unplug_challenge",
      +  "abandon_unplug_challenge",
      +  "merge_tags",
      +  "favorite_memory",
      +  "remind_witness",
      +  "undo_challenge_contribution",
      +  "mark_learning_volumes",
      +  "duplicate_workout_template",
      +  "link_items",
      +  "send_friend_request",
      +  "set_budget_limit",
      +  "set_focus_goal",
      +  "set_screen_time_goal",
      +  "skip_task_occurrence"
      +]
  2. First observed

TDQS

A4.4/5.0
Behavior5/5

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

Beyond the destructiveHint=true annotation, the description discloses the specific destruction each action causes: merge_tags makes the source tag disappear, abandon_unplug_challenge is explicitly irreversible ('sans retour possible'), and mark_learning_volumes silently creates a volume when a number exceeds what is published. It also documents a rate limit, paywall gating, and the subtle counting behaviour of skip_task_occurrence in a 'après N fois' completion – all information the annotations cannot express.

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

Conciseness4/5

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

The general rule is front-loaded before the enumeration, and the bullet-per-action format makes 18 operations scannable; almost no sentence is filler. It loses a point because several bullets are run-on, packing field semantics, cross-references and edge cases into a single dense clause, which is harder to parse in a hurry.

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?

For a single entry point behind 18 heterogeneous mutations, the definition covers what each action does, what it needs, and what it destroys, and it points to siblings for follow-up reads since no output schema exists. Remaining gaps are minor: it does not say what a successful call returns, and for target-less actions such as set_focus_goal or set_screen_time_goal it never clarifies that `id` is irrelevant.

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

Parameters5/5

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

Schema coverage is 100%, but the `fields` object is free-form with only 'voir la liste' as its schema description, so the description carries the real burden and does: it names the keys per action (target_id, direction, used_at, from/to/owned/read/current, budget_category_id/month/amount, minutes, daily_limit_minutes, from_type/from_id/to_type/to_id) with accepted values and defaults. It also spells out the per-action meaning of parent_id, which the schema only partially echoes.

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

Purpose4/5

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

The opening sentence defines a real scope boundary: an action that is neither a creation nor a field update, which distinguishes it from the create/update siblings. The 18 bullets then enumerate exactly which operations qualify, so the tool's remit is concrete rather than tautological despite the vague name 'faire une action'.

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?

Each action carries its own activation conditions and prerequisites: plan Plus required for advance_key_result, remind_witness and duplicate_workout_template, 'un seul à la fois' for start_unplug_challenge, a pseudo needed before send_friend_request, and a 48-hour limit stated for remind_witness. The definition also routes the agent to siblings for related reads (dandori_unplug_dashboard, fetch learning:ID, dandori_delete entity link), though it never states outright when to prefer a dedicated sibling over this dispatcher.

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.

Resources