Skip to main content
Glama

InProcess Cockpit

Lancer une commande

commande
Destructive

Lance une commande dans le dossier du cockpit, parmi celles de l'agent du Studio : pnpm exec coulisse codegen | check | verify --fast | test tests/x.test.ts | db generate | db migrate | db seed --reset, pnpm add , pnpm exec tsc --noEmit, git status | diff | log, ls, grep… (6 min au plus).

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": {
      "cockpit": "<cockpit>",
      "commande": "<commande>"
    },
    "name": "commande"
  }
}

Input Schema

NameRequiredDescriptionDefault
cockpitYesIdentifiant du cockpit (mes_cockpits)
commandeYes

Schema Changelog

Changes observed during successful MCP inspections.

No schema history has been recorded yet.

TDQS

A4/5.0
Behavior4/5

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

Annotations already flag readOnlyHint=false and destructiveHint=true, so safety is covered; the description adds genuinely new behavioral facts: execution happens in the cockpit folder, commands are limited to the Studio agent's whitelist, and there is a hard "6 min au plus" timeout. It does not disclose what is returned (stdout/stderr), truncation, or what happens on non-zero exit.

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?

Front-loaded with purpose and location before the enumeration, and the timeout is appended at the end where it is easy to scan. The command list is long and slightly dense with pipe separators, but each entry carries real information about what is allowed.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a destructive execution tool with no output schema, the definition covers location, permitted commands, and timeout — the essentials for calling it safely. It leaves a meaningful gap around results: no indication of how output is returned, how errors surface, or whether long output is truncated, which matters for a tool whose entire product is command output.

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

Parameters3/5

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

Schema coverage is 50%: the `cockpit` parameter is documented in-schema ("Identifiant du cockpit (mes_cockpits)") while `commande` has no schema description at all. The description compensates well for `commande` by enumerating acceptable command forms and syntax examples, but adds nothing about the `cockpit` identifier beyond what the schema already says.

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 ("Lance une commande") plus the execution context ("dans le dossier du cockpit") and enumerates the exact command families permitted. The whitelist of pnpm/git/ls/grep commands makes it unmistakable that this is the shell-execution tool rather than the sibling file tools (lire_fichier, ecrire_fichier).

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?

Constrains usage explicitly with "parmi celles de l'agent du Studio" and lists representative allowed commands (codegen, check, verify --fast, test, db generate/migrate/seed, tsc, git status/diff/log, ls, grep), which tells the agent what belongs here. It stops short of naming when to prefer a sibling (e.g., use lire_fichier instead of cat) or stating exclusions beyond the implicit whitelist.

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