list_projectsOpen the native Project OS entry point and list recent 2ools projects the authenticated account can access. The app automatically continues the only saved project or shows an inline chooser when several projects are available. Returns compact metadata without generated source.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Maximum projects to return. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, covering safety. The description adds useful behavioral context beyond annotations: it opens the native Project OS entry point, auto-continues the only saved project, shows an inline chooser when multiple are available, and returns compact metadata without generated source. This enriches the agent's understanding of what happens on invocation.
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 three sentences with no extraneous content. It front-loads the core purpose and adds concise behavioral notes. The phrase 'Open the native Project OS entry point' is somewhat redundant but not harmful; overall it is well-structured and efficient.
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?
With no output schema, the description appropriately explains the return value ('compact metadata without generated source'). It covers key context: auth scope, behavior for one vs. multiple projects, and the read-only nature. Nothing critical for an agent to call this tool 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 documents the sole 'limit' parameter with a clear description ('Maximum projects to return') and 100% coverage. The description adds no additional details about the parameter (e.g., ordering, defaults) beyond what the schema provides. Baseline of 3 is appropriate given complete schema coverage.
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 clearly states the verb 'list' and the resource 'recent 2ools projects' accessible to the authenticated account. It distinguishes from siblings like 'get_project' by focusing on listing multiple projects and explicitly notes that it returns compact metadata without generated source, which further differentiates it from tools that return full project data.
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 implies this is the entry point for browsing projects, but it does not explicitly explain when to use it versus alternatives like 'continue_project' or 'activate_project'. The behavior of auto-continuing or showing a chooser suggests it is for selection, but no direct comparison or exclusion is provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
With 74 tools, many serve similar project/version/agent workflows (e.g., get_project vs get_project_context vs get_project_activity). Detailed descriptions help but the sheer volume creates selection risk for agents, especially between approve_version/approve_artifact_version and reject counterparts.
Names predominantly follow a verb_object pattern (list_projects, create_public_preview, update_project_mission). Exceptions like 'act_on_project_os_task' and 'build_from_conversation' are minor, but the strong convention overall makes it predictable.
74 tools is extreme—more than double the typical upper bound and far beyond what an agent can feasibly consider per task. The domain is broad, but the count borders on overwhelming and suggests lack of consolidation.
The surface covers project lifecycle, versioning, builds, agents, artifacts, feedback, and work assignments comprehensively. Yet there is no explicit create/delete project tool (only via build_from_conversation or start_project_on_engine), and no direct update to project metadata beyond mission/plan, leaving minor gaps.