Official MCP RegistryListed
Unseen Workflow Compiler
Compile SOPs, policies, and playbooks into source-grounded visual flowcharts (SVG, Mermaid, JSON).
First seen 22 Oct 2026. Evidence as of 14 Nov 2026.
11
Tools
From an anonymous probe
1
Source listings
Each with its own history
19
Recorded changes
Since first seen
Tools
| Tool | Description | Behaviour |
|---|---|---|
| auth_login | Legacy sign-in for an existing password account. Account creation and password setup require verified browser sign-in. ChatGPT/Codex plugin connections use browser OAuth instead of this tool. | Changes data |
| auth_request_otp | Step 1 of Agent & Human Login (Email + OTP): Sends a 6-digit Login OTP (`UNSEEN_LOGIN_OTP`) to the user's email address. This is a passwordless Login OTP (NOT a password reset). Once called, the agent can either read the 6-digit `UNSEEN_LOGIN_OTP` directly from the user's email (when the user has granted email access) or ask the user in chat for the 6-digit code, then call `auth_verify_otp`. | Changes data |
| auth_verify_otp | Step 2 of Agent & Human Login (Email + OTP): Verifies the 6-digit Login OTP sent to `email` and returns `{ token: 'utok_...', user }` (30 free credits / 3 visuals per month on new accounts). Pass the returned `token` as `Authorization: Bearer <token>` or in the `authToken` argument of subsequent MCP tool calls. | Changes data |
| autofix_workflow | Deterministic connector healing and full visual re-rendering ($0 compute cost, instant, unlimited). Cleans exact duplicate edges and empty sequence labels, recomputes deterministic orthogonal layout, and returns fresh `spec`, `svg`, `mermaid`, `excalidraw`, and `studioUrl`. Missing steps or branches require source-backed edits to the spec; this tool preserves graph meaning and does not perform semantic verification. Use this tool after making direct structural edits to `VisualSpec` JSON (`nodes` / `edges`), or when a user asks to autofix connectors. | Changes data |
| compile_workflow | Turn supplied document evidence, raw operational text, OR a Host-LLM VisualSpec v0.1 JSON (`spec`) into a publication-ready workflow diagram with source references, clean vector SVG, Mermaid, Excalidraw scene, and interactive Studio deep link (`studioUrl`). THREE INPUT MODES: (1) HOST-LLM NATIVE MODE ($0 free, unlimited once signed in, <50ms — ALWAYS PREFER THIS WHEN YOU HAVE A HOST LLM): Construct `spec` (VisualSpec v0.1 JSON: `{schemaVersion:'0.1', kind:'flowchart', title, summary, sources:[{id,locator,excerpt}], nodes:[{id,label,role:'step'|'decision'|'concept',actor?,detail?,sourceIds:[...]}], edges:[{from,to,label?,sourceIds:[...]}]}` where outgoing edges from `decision` nodes MUST have a `label`) using your own LLM and pass `{ spec }`. Unseen deterministically validates citations, heals connectors, and renders SVG/Mermaid/Excalidraw/studioUrl at $0 Unseen credits even when server-AI credits are 0 (returning exact validation errors if anything needs self-correction). (2) RAW TEXT MODE (PAID / USES ALLOWANCE): Passing ONLY `{ text: '...' }` without `spec` uses Unseen's server-side AI pipeline and returns HTTP 402 if free monthly slots/credits are 0 — if that happens, switch to Mode 1 (`{ spec }`) for $0. (3) STRUCTURED SOURCES MODE (PAID / USES ALLOWANCE): Passing ONLY `{ sources: [{id, locator, excerpt}, ...] }` without `spec` uses Unseen's server-side AI pipeline (requires free monthly slot or credits). EDITING GUIDANCE: For cosmetic edits, edit `<svg>` XML directly. For structural edits, edit `spec` JSON and call `compile_workflow({ spec })` or `autofix_workflow({ spec })` ($0 free). | Changes data |
| detect_workflows | Discover distinct operational workflows from multi-section documents or raw text before compiling. Returns candidate workflows (`id`, `title`, `description`, `sourceIds`) audited by TypeSafe Jev. | Read-only |
| enhance_workflow | Bundled free semantic verification and repair powered by TypeSafe Jev. Audits the VisualSpec against its cited `sources[]` for unsupported claims, missing decision branches, and role accuracy, applies targeted repair if actionable issues are found, and returns the enhanced `spec`, `svg`, `mermaid`, `excalidraw`, `studioUrl`, and Jev quality metrics. | Changes data |
| get_account_status | Check the authenticated user's Unseen credit balance, free monthly server-AI generation allowance, uniform 100-page safety limit per visual (Host-LLM spec compilation uses 0 credits), and 1-click prepaid credit refill link (`https://unseendocs.in/#pricing`). Call this before server-AI compilations or when credits are low. | Read-only |
| get_workflow | Retrieve an existing VisualSpec workflow by jobId or visualId, returning the spec, SVG layout, and Mermaid diagram. | Read-only |
| revise_workflow | Apply explicit node label/role changes or node removal using node IDs, or request connector autofix. For other changes, including new steps or edges, edit the returned spec with confirmed source evidence and call autofix_workflow. Unsupported instructions return an error; this tool does not perform general natural-language rewrites. | Changes data |
| send_feedback | Report a bug, a missing capability, or an improvement idea about Unseen's tools or output to the Unseen team. No sign-in needed and free. Use it whenever a tool errors, behaves unexpectedly, or you had to work around something. Describe the problem in your own words (what you called, what you expected, what happened). Never include document contents, personal data or credentials. | Changes data |
Change history
- send_feedback: tool added
- revise_workflow: input schema changed (+includeExcalidraw, +includeSvg, +maxPreviewHeight, +maxPreviewWidth, +renderFormat)
- get_workflow: input schema changed (+includeExcalidraw, +includeSvg, +maxPreviewHeight, +maxPreviewWidth, +renderFormat)
- get_account_status: tool added
- enhance_workflow: input schema changed (+includeExcalidraw, +includeSvg, +maxPreviewHeight, +maxPreviewWidth, +renderFormat)
- detect_workflows: input schema changed (+evidenceCoverage, +granularity)
- compile_workflow: input schema changed (+evidenceCoverage, +focusSourceIds, +includeExcalidraw, +includeSvg, +maxPreviewHeight, +maxPreviewWidth, +renderFormat, +spec)
- compile_workflow: description changed (+"evidence," +"text, OR a Host-LLM VisualSpec v0.1 JSON (`spec`)" +"THREE")
- autofix_workflow: input schema changed (+includeExcalidraw, +includeSvg, +maxPreviewHeight, +maxPreviewWidth, +renderFormat)
- server instructions changed (+"Unseen Workflow Compiler turns SOPs, compliance policies, spreadsheets, playbooks, and multi-actor processes into source-grounded workflow visuals (clean SVG, Mermaid, Excalidraw, and interactive Studio links). OPERATING RULES: 1. Prefer Host-LLM Native Mode ($0 Unseen credits, <50ms): Read the full document in your own context first, synthesize a bounded VisualSpec v0.1 JSON (`{ schemaVersion: '0.1', kind: 'flowchart', title, summary, sources: [{ id, locator, excerpt }], nodes: [{ id, label, role: 'step'|'decision'|'concept', actor?, detail?, sourceIds }], edges: [{ from, to, label?, sourceIds }] }`), and call `compile_workflow({ spec })`. Only pass raw `{ text }` or `{ sources }` when you explicitly want Unseen's server-side AI pipeline (which uses monthly allowance/credits). 2. Inspect embedded diagrams & tables before claiming missing steps: Copy diagram step order and edges into their own cited source excerpt with locator (e.g. `p.9 figure: process map`). 3. Stay source-grounded: Preserve documented rejection paths, exceptions, approvals, thresholds, and timing. Every outgoing edge from a `decision` node MUST carry a source-backed `label`. Never invent missing branches; surface ambiguity or contradictions cleanly. 4. Present & self-heal: Render the returned `svg` / `svgUrl` / `mermaid` directly in chat alongside the optional `studioUrl`. Fix structural or connector warnings yourself via `autofix_workflow`, `revise_workflow`, or `compile_workflow({ spec })` ($0 credits). 5. Built-in Skill & Contract: Read MCP resources `unseen://skills/unseen-workflow-compiler/SKILL.md` and `unseen://contract/visual-spec-v0.1` for the full workflow skill and schema bounds, or invoke built-in prompts (`compile-sop-host-llm`, `audit-workflow-gaps`, `revise-workflow-from-feedback`). Call `send_feedback` (no auth required, never include raw docs or PII) if a tool errors or behaves unexpectedly.")
- compile_workflow: input schema changed
- autofix_workflow: input schema changed (+swimlane)
- autofix_workflow: description changed (+"Cleans exact duplicate edges and empty sequence labels," +"Missing steps or branches require source-backed edits to the spec; this tool preserves graph meaning and does not perform semantic verification." -"Rescues orphan nodes, heals broken or missing decision branches,")
- autofix_workflow: annotations changed
- revise_workflow: description changed (+"explicit node label/role changes or node removal using" +"IDs, or request connector autofix. For other changes, including new steps or" +"edit the returned spec with confirmed source evidence")
- compile_workflow: description changed (+"INPUT SCOPE: Use reviewed written procedures only. Do not interpret scans, images, chart/diagram geometry or unlabeled arrows, and do not certify business procedures. Ask for reviewed written process text or source-backed clarification.")
- auth_login: input schema changed
- auth_login: description changed (+"Legacy sign-in for" +"password account. Account creation" +"password setup require verified browser sign-in. ChatGPT/Codex plugin connections")
- Listed (registry)
| Source | Listing | First seen | Last seen | Versions |
|---|---|---|---|---|
| Official MCP Registry | io.github.eyeinthesky6/unseen-workflow-compiler | 22 Oct 2026 | 14 Nov 2026 | 1 |