dokploy_compose

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

With no annotations provided, the description carries the full disclosure burden and excels at it: get 'never values' (masked summary), setEnvVars 'merge inside the server, masked confirmation only', saveEnvironment 'full replace', getEnvValuesUnsafe explicitly flagged 'UNSAFE' with output going to the tool transcript, and update/createEnvFile's side effect of writing a .env file beside the compose file. It also reveals the latent first-deploy failure mode and the search index's narrowness — exactly the behavioral knowledge an agent cannot derive from the schema.

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?

For a 21-action dispatcher, the description is remarkably dense with zero filler — each action is compressed into a signature plus caveats, and the most important safety warnings are front-loaded within each segment. The weakness is purely structural: a single unbroken paragraph that is hard to scan; per-action line breaks or a bulleted layout would make it exemplary.

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 39 parameters, 21 actions, no annotations, and no output schema, the description covers the high-risk and high-confusion paths well: the unsafe env escape hatch, masking behavior, deployment retry semantics, full-replace vs merge distinctions, and a cross-tool pointer (dokploy_docker findContainers for readLogs discovery). The tail actions (cancelDeployment/cleanQueues/killBuild/refreshToken) are name-listed without any behavioral explanation, and return shapes are only specified for env-related and get actions, leaving other actions' outputs unspecified.

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 64% (mid-range), and the description's main contribution is the action→parameter routing — critical for a tool where only 'action' is required and the other 38 params are context-dependent. It adds real semantics beyond the schema ('createEnvFile? (also write a .env file next to the compose file)', serviceNetworks per-entry requirements, env as KEY=VALUE lines). However, it references 'composeFile' and 'containerId' which do not map cleanly to the schema (containerId is documented under the oddly-named 'description' property; composeFile doesn't exist as a property despite additionalProperties:false).

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 opening clause 'Manage Docker Compose services' plus the exhaustive action enumeration (create/get/update/deploy/delete/start/stop/move/readLogs/search, etc.) makes the tool's scope fully unambiguous. It stays at 4 rather than 5 because the verb 'Manage' is generic and it never explicitly names a sibling it is not (e.g., dokploy_application), relying instead on the 'compose' resource to differentiate.

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 embeds per-action usage constraints throughout, e.g., 'loadServices: composeId (must deploy first)', deploy's 'first deploy on new services may fail — retry immediately', and the search caveat that the API index is narrower than the project tree, explicitly recommending dokploy_project/dokploy_overview for inventory instead. However, there is no global when-to-use-vs-alternatives statement for the tool itself or for most actions; the agent must infer applicability from the action signatures.

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