planexe.org

PlanExe generates strategic project-plan drafts from a natural-language prompt. Output is a self-contained interactive HTML report (~700KB) with 20+ sections including executive summary, interactive Gantt charts, risk analysis, SWOT, governance, investor pitch, team profiles, work breakdown, scenario comparison, expert criticism, and adversarial sections (premortem, self-audit checklist, premise attacks) that stress-test whether the plan holds up. The output is a draft to refine, not final ground truth — but it surfaces hard questions the prompter may not have considered. Use PlanExe for substantial multi-phase projects with constraints, stakeholders, budgets, and timelines. Do not use PlanExe for tiny one-shot outputs (for example: 'give me a 5-point checklist'); use a normal LLM response for that. The planning pipeline is fixed end-to-end; callers cannot select individual internal pipeline steps to run. Required interaction order: call example_plans first (optional, to preview what PlanExe output looks like — curated example reports and zip bundles). Then call example_prompts. Optional before plan_create: call model_profiles to see profile guidance and available models in each profile. Then perform a non-tool step: draft a strong prompt as flowing prose (not structured markdown with headers or bullets), typically ~300-800 words, and get user approval. Good prompt shape: objective, scope, constraints, timeline, stakeholders, budget/resources, and success criteria. Write the prompt as flowing prose — weave specs, constraints, and targets naturally into sentences. Only after approval, call plan_create. Each plan_create call creates a new plan_id; the server does not enforce a global per-client concurrency limit. Then poll plan_status (about every 5 minutes) to check progress; use plan_file_info when complete. plan_status is the primary way to track progress — it returns structured JSON with progress_percentage, steps_completed/steps_total, and current_step. Optionally, run `curl -N <sse_url>` in a background shell as a completion detector — the stream auto-closes when the plan reaches a terminal state (completed/failed/stopped). If plan generation fails before completing all steps, call plan_resume to continue from where it left off without discarding completed work. Use plan_retry instead for a full restart (plan must be in failed state). Both accept a failed plan_id and optional model_profile (defaults to baseline). To stop, call plan_stop with the plan_id from plan_create; stopping is asynchronous and the plan will transition to the stopped state. If model_profiles returns MODEL_PROFILES_UNAVAILABLE, inform the user that no models are currently configured and the server administrator needs to set up model profiles. Tool errors use {error:{code,message}}. plan_file_info returns {ready:false,reason:...} while the artifact is not yet ready; check readiness by testing whether download_url is present in the response. plan_file_info download_url is the absolute URL where the requested artifact can be downloaded. To list recent plans for a user call plan_list; returns plan_id, state, progress_percentage, created_at, and prompt_excerpt for each plan. plan_status state contract: pending/processing => keep polling; completed => download is ready; failed => terminal error; stopped => user called plan_stop (consider plan_resume). Troubleshooting: if plan_status stays in pending for longer than 5 minutes, the plan was likely queued but not picked up by a worker (server issue). If plan_status is in processing and output files do not change for longer than 20 minutes, the plan_create likely failed/stalled. In both cases, report the issue to PlanExe developers on GitHub: https://github.com/PlanExeOrg/PlanExe/issues . Main output: a self-contained interactive HTML report (~700KB) with collapsible sections and interactive Gantt charts — open in a browser. The zip contains the intermediary pipeline files (md, json, csv) that fed the report. Use send_feedback to report issues or share observations about plan quality, workflow friction, or the MCP interface. Feedback is fire-and-forget and never blocks the workflow. New users: create an account and obtain an API key at https://home.planexe.org/ .