MCP server for Agendo, an AI agent orchestration platform. Provides tools for artifact rendering, brainstorming, planning, progress tracking, project management, and session control.
Agendo provides 10 tools with explicit schemas and descriptions. Most tools have clear verb-noun naming (render_artifact, brainstorm_signal, save_plan, add_progress_note). Descriptions are present and contextual (avg ~120 chars), meeting baseline minimums. However, several tools lack parameter descriptions or have incomplete schemas. Error handling guidance is minimal, tools do not explain recovery paths or categorize failures. Output schemas are not documented. Some parameters (e.g., 'reason' in brainstorm_signal) lack type constraints. The server demonstrates solid foundational quality but falls short of production-grade polish in error messaging and output documentation.
Add a progress note to the current task. Use this to report milestones, blockers, or intermediate results.
Assign a task to an agent by slug
Get current brainstorm room state: wave number, participants, who has responded, your role, and whether you are the leader.
Signal your brainstorm turn state. Use instead of text-based [PASS]. Call with done when your response is complete, pass when you agree and have nothing to add (reason required), or block for a critical objection (reason required).
Get the progress note history for a task. Returns all agent_note events in chronological order.
Get the full details of a project by its UUID
No output schemas documented. Tools return data but LLMs cannot predict field structure, forcing runtime parsing and risking context loss.
Error handling lacks recovery guidance. Tools do not explain what to do on failure (retry, ask user, abort). Agents cannot self-correct.
Parameter 'reason' in brainstorm_signal lacks type constraint (should be string with minLength). Parameter 'permissionMode' in start_agent_session has enum but no description of what each mode does.
brainstorm_get_state and list_projects have minimal descriptions (60-65 chars). Descriptions should explain WHEN to call and WHAT structure is returned.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | 2026-07-28+ | v2 |
List all projects. By default returns only active projects.
Render an interactive visual (chart, dashboard, diagram, UI mockup, SVG) inline in the Agendo chat. Use when a visual communicates better than text. Design guidelines are pre-loaded via the artifact-design skill — apply them when writing HTML/SVG.
Save or update an implementation plan. Call this when your plan is finalized and ready for review. The plan will be linked to the current project.
Start a new agent session for a task. The session runs asynchronously (fire-and-forget). Returns a sessionId you can use to monitor progress.
Pagination not visible in list_projects. If results can exceed ~20 items, tool should accept limit/offset and return total_count or next_cursor.