Forge Engine
Design spec + milestones AI coding agents read before building; drift flagged, changes reviewed.
我该使用它吗
质量与安全性
发现(19)
- HIGH在 get_build_region 中
- HIGH
- MEDIUM在 report_build_status 中
- MEDIUM在 report_screen_status 中
- LOW在 list_projects 中
- LOW在 get_system 中
- LOW在 list_systems 中
- LOW在 list_screens 中
- LOW在 reorder 中
- LOW在 list_milestones 中
基于对工具定义和协议合规性的自动分析。
上下文开销
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中:
{
"mcpServers": {
"forge": {
"url": "https://mmvdabzadclebfxyzudg.supabase.co/functions/v1/forge-mcp"
}
}
}远程端点
https://mmvdabzadclebfxyzudg.supabase.co/functions/v1/forge-mcpstreamable-http它能做什么
工具清单
工具(54)
🟢get_help
Forge how-to on demand (no project needed): propose→adopt, what a System is, Balance Lab formats, sign-in. Call when unsure.
输入模式
{
"type": "object",
"properties": {}
}🟢get_workflow
Run FIRST in a new repo (no project): first-connect routine + the working-agreement block to ADD to AGENTS.md/CLAUDE.md now (tell the user in one line; remove if they object).
输入模式
{
"type": "object",
"properties": {}
}🟢import_from_code(project_id)
One-command codebase import: recipe + batch id — propose one Context overview + one System per real module; the owner adopts all at once.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢design_ui_from_systems(project_id)
Recipe to design the UI from Systems (user request only): steps + batch id — confirm canvas aspect, then screens + PLACED elements + edges in one batch.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟡post_log(text, entity, version, project_id)
Append a build-log entry to Activity — what you built/decided (commit-note style).
输入模式
{
"type": "object",
"properties": {
"text": {
"type": "string",
"description": "What you did / decided (1–2 sentences)"
},
"entity": {
"type": "string",
"description": "System/screen/milestone it's about"
},
"version": {
"type": "string",
"description": "e.g. \"0.379.0\""
},
"project_id": {
"type": "string"
}
},
"required": [
"text"
]
}🟢get_project_context(project_id)
FULL design dump — LARGE, last resort; prefer get_project_meta + list_*/get_*/search.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢list_projects
Projects your key reaches (id, name, role); pass an id as project_id to switch.
输入模式
{
"type": "object",
"properties": {}
}🟢get_project_meta(project_id)
START HERE. Tiny overview: counts, members, task claims + working_now (avoid collisions), forge_workflow_version.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢trace_ui_from_image(screen, project_id)
SECONDARY (user-driven UI): turn a screen's REFERENCE IMAGE into placed elements. Returns the image itself plus the recipe — the canvas resolution, the fraction→pixel conversion that stops coordinates landing wrong, and what is already on the screen so a second pass updates instead of duplicating. You look at the picture and send back propose_element / update_element with x/y/w/h; it all lands in the owner's Inbox. Use when the user wants the wireframe to match a screenshot or mockup they uploaded.
输入模式
{
"type": "object",
"properties": {
"screen": {
"type": "string",
"description": "The screen whose reference image to trace (name or id)"
},
"project_id": {
"type": "string"
}
},
"required": [
"screen"
]
}🟢get_stale(project_id)
What has drifted between the Idea lane and the System Specs — the design's own out-of-sync list, computed deterministically (no AI, no tokens). Three kinds: Idea notes edited since the systems were generated from them (with the systems each one touches), Idea notes whose prose the systems have moved past, and specs whose stamped source fingerprint no longer matches. Read this before assuming the design is coherent; fix the first kind with resync_from_idea.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢resync_from_idea(note, project_id)
The recipe for re-syncing ONE changed Idea note into the specs that depend on it — the same scoped job the app's "Re-sync N systems" button does, minus the button. Returns the note, the affected specs in full, and exactly how to send the result back. You do the writing; it lands in the owner's Inbox to ADOPT. CRUCIAL: send updates with `resync:true`, or the owner adopts your fix and the design still reports it as stale.
输入模式
{
"type": "object",
"properties": {
"note": {
"type": "string",
"description": "The Idea note's title or id (from get_stale)"
},
"project_id": {
"type": "string"
}
},
"required": [
"note"
]
}🟢get_design_document(for_summary, page, project_id)
The WHOLE design as one readable document — Vision (+ Project DNA) → every System spec → reference notes, compiled deterministically from the current design. Read this to understand a project end-to-end instead of walking list_systems → get_system N times. Returns markdown plus the project `version` it was compiled from. Long designs come back PAGED — the header says 'part N of M', call again with `page: N+1` for the rest. Pass `for_summary: true` to get the condensed projection instead (every system's Goal + Boundary, tables stripped, one page) — that is what you should summarize from.
输入模式
{
"type": "object",
"properties": {
"for_summary": {
"type": "boolean",
"description": "Return the condensed summarization source (all system goals + boundaries, no tables) instead of the full document"
},
"page": {
"type": "number",
"description": "Which part of the full document to return (default 1). The header tells you how many parts there are."
},
"project_id": {
"type": "string"
}
}
}🔴set_design_overview(summary, project_id)
Write the Design Document's Overview — the human-readable page a new team member reads first. WRITES DIRECTLY (no Inbox): it is a derived, clearly-labelled AI summary, not design truth, and the owner can clear or rewrite it in one click. HARD RULES, same as the in-app button: use ONLY facts stated in the design (call get_design_document with for_summary:true first); invent no mechanics, numbers or names; describe, never evaluate; write in the design's dominant language. Structure: `### What this is` · `### The core loop` · `### How the systems fit` (which system feeds which — the part a raw spec list cannot give) · `### Edges` (ONLY if the design states scope limits / open questions). 250-400 words, no top-level heading. Forge stamps the project version it was compiled from, so the owner is told when the design has moved past it.
输入模式
{
"type": "object",
"properties": {
"summary": {
"type": "string",
"description": "The Overview in markdown, starting at `### What this is`"
},
"project_id": {
"type": "string"
}
},
"required": [
"summary"
]
}🟢get_briefing(files, task, project_id)
ONE-CALL orientation before you build: pass `files` you're about to edit (or a `task`) → the systems that own them, each with Goal + Acceptance + build status/files/last_commit/drift + coupled_systems (code neighbors an edit may break) + pending Inbox changes + recent decisions, plus open-rejection count, who's active, `recent_changes` (what happened since you were last here — a session handoff), and a task to claim. Replaces the get_project_meta→get_system→get_build_region dance.
输入模式
{
"type": "object",
"properties": {
"files": {
"type": "array",
"items": {
"type": "string"
},
"description": "Repo-relative paths you're about to touch — mapped to their systems"
},
"task": {
"type": "string",
"description": "A task id or name — briefs its milestone's systems instead"
},
"project_id": {
"type": "string"
}
}
}🟢get_system(id, name, project_id)
ONE system's full spec (Goal/Boundary/Acceptance, markdown, sources, status) + pending Inbox changes touching it.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"project_id": {
"type": "string"
}
}
}🟢list_systems(project_id)
All systems, compact: id, name, status, 1-line goal.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢get_build_region(system, page, project_id)
System→code map per built system: implementing files, status, drift flag, last_commit, and acceptance-evidence COUNTS. Mapped files gone from the repo? report_drift. Pass `system:"<name|id>"` for ONE system plus the full text of its acceptance criteria (omitted from the map — it is the bulk of the payload). Big projects come back paged: the body says 'page N of M', call again with `page: N+1`.
输入模式
{
"type": "object",
"properties": {
"system": {
"type": "string",
"description": "One system by name or id — returns its acceptance-criteria detail too"
},
"page": {
"type": "number",
"description": "Which page of the map (default 1); the body tells you how many there are"
},
"project_id": {
"type": "string"
}
}
}⚪report_drift(id, name, missing_files, note, project_id)
Flag CODE DRIFT — mapped files no longer match the repo. Advisory; a fresh report_build_status clears it.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"missing_files": {
"type": "array",
"items": {
"type": "string"
},
"description": "Paths that no longer exist"
},
"note": {
"type": "string",
"description": "Optional — what drifted"
},
"project_id": {
"type": "string"
}
}
}🟢get_history(system, query, limit, project_id)
DESIGN MEMORY: recorded decisions/logs/rejections with who/when. Read BEFORE changing a system's direction; empty = no recorded WHY — don't invent one.
输入模式
{
"type": "object",
"properties": {
"system": {
"type": "string",
"description": "System id or exact name"
},
"query": {
"type": "string",
"description": "Keyword filter"
},
"limit": {
"type": "number",
"description": "Default 30, max 100"
},
"project_id": {
"type": "string"
}
}
}🟢get_impact(id, name, project_id)
BLAST RADIUS of a system (deterministic): upstream context, siblings, dependent screens/milestones, files, code-coupled systems, pending Inbox, recent activity. Run BEFORE changing it.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"project_id": {
"type": "string"
}
}
}🟢get_rejections(project_id)
Owner's DECLINED list — check at session start; follow each entry's guidance, then resolve_rejection(title).
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟡resolve_rejection(title, project_id)
Report you REVERTED a declined change (unlocks the owner's Clear). Call after realigning the build.
输入模式
{
"type": "object",
"properties": {
"title": {
"type": "string",
"description": "Exactly as get_rejections returned it"
},
"project_id": {
"type": "string"
}
},
"required": [
"title"
]
}🟢list_screens(project_id)
All screens, compact: id, name, purpose, status.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢get_screen(id, name, project_id)
ONE screen's layout: canvas `resolution` (use THESE px), elements x/y/w/h, links, popups. Read before editing a screen. has_reference_image:true → get_screen_image shows you the actual image.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"project_id": {
"type": "string"
}
}
}🟢get_screen_image(id, name, project_id)
A screen's reference image (HUD background) as an actual IMAGE you can see — reads the stored file inline (signed-URL fallback past 4MB).
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"project_id": {
"type": "string"
}
}
}🟡update_element(element_id, screen, label, x, y, ...)
Move/resize/relabel an element — DIRECT, live. element_id (preferred) or screen+label; x+y also places an unplaced one.
输入模式
{
"type": "object",
"properties": {
"element_id": {
"type": "string",
"description": "From get_screen (preferred)"
},
"screen": {
"type": "string",
"description": "With label, if no element_id"
},
"label": {
"type": "string",
"description": "Current label"
},
"x": {
"type": "number",
"description": "px on the screen's canvas"
},
"y": {
"type": "number"
},
"w": {
"type": "number"
},
"h": {
"type": "number"
},
"new_label": {
"type": "string",
"description": "Rename (optional)"
},
"type": {
"type": "string",
"description": "e.g. button/bar/panel (optional)"
},
"note": {
"type": "string",
"description": "\"\" clears"
},
"system": {
"type": "string",
"description": "System it serves"
},
"project_id": {
"type": "string"
}
}
}🔴withdraw_proposal(id, project_id)
Remove YOUR OWN still-pending Inbox item (id from the propose response / get_inbox).
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "The pending item's id"
},
"project_id": {
"type": "string"
}
},
"required": [
"id"
]
}🔴delete_entity(kind, id, name, from, to, ...)
PROPOSE a delete → Inbox (owner adopts; nothing deleted now). id preferred or exact name; flow_edge may use from+to.
输入模式
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"system",
"context",
"screen",
"milestone",
"task",
"element",
"flow_edge"
],
"description": "What to delete"
},
"id": {
"type": "string",
"description": "Entity/task/element/edge id"
},
"name": {
"type": "string",
"description": "Exact name/title (if no id)"
},
"from": {
"type": "string",
"description": "flow_edge: source screen"
},
"to": {
"type": "string",
"description": "flow_edge: destination screen"
},
"project_id": {
"type": "string"
}
},
"required": [
"kind"
]
}🔴dedupe(kind, screen, project_id)
Remove duplicate-named entries (keep first) — DIRECT, destructive.
输入模式
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"milestones",
"elements"
],
"description": "What to dedupe"
},
"screen": {
"type": "string",
"description": "For kind=elements"
},
"project_id": {
"type": "string"
}
},
"required": [
"kind"
]
}⚪reorder(kind, order, milestone_id, milestone, screen, ...)
Reorder milestones / tasks / elements — DIRECT. order = ids in new order; omitted keep relative order.
输入模式
{
"type": "object",
"properties": {
"kind": {
"type": "string",
"enum": [
"milestones",
"tasks",
"elements"
],
"description": "What to reorder"
},
"order": {
"type": "array",
"items": {
"type": "string"
},
"description": "Ids in the new order"
},
"milestone_id": {
"type": "string",
"description": "For kind=tasks"
},
"milestone": {
"type": "string",
"description": "For kind=tasks (name)"
},
"screen": {
"type": "string",
"description": "For kind=elements"
},
"project_id": {
"type": "string"
}
},
"required": [
"kind",
"order"
]
}🟢list_milestones(project_id)
All milestones, compact: id, name, weeks, order, done/total.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢get_milestone(id, name, project_id)
ONE milestone: goal, weeks, difficulty, systems, every task (id/name/done/status/effort).
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"project_id": {
"type": "string"
}
}
}🟢list_activity(query, kind, who, limit, project_id)
Change history {who, change, entity, kind, when}, newest first; filter kind/who/query.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Keyword filter (optional)"
},
"kind": {
"type": "string",
"description": "system | context | milestone | screen (optional)"
},
"who": {
"type": "string",
"description": "Author email substring (optional)"
},
"limit": {
"type": "number",
"description": "Default 30, max 200"
},
"project_id": {
"type": "string"
}
}
}🟢get_inbox(project_id)
PENDING Inbox (the owner's triage queue). Check BEFORE proposing — avoid duplicates. Read-only.
输入模式
{
"type": "object",
"properties": {
"project_id": {
"type": "string"
}
}
}🟢search(query, kinds, project_id)
Keyword search — compact hits {kind, id, name} + snippet.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Text to search for"
},
"kinds": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional: system | context | milestone | task | screen | proposal"
},
"project_id": {
"type": "string"
}
},
"required": [
"query"
]
}🟡propose_system(title, spec, files, derives_from, batch, ...)
Propose a NEW system → Inbox. spec = ## Goal / ## Boundary (Owns · Doesn't own) / ## Acceptance. Exists? use update_system. (Alias: create_proposal.)
输入模式
{
"type": "object",
"properties": {
"title": {
"type": "string",
"description": "System name (Title Case, English)"
},
"spec": {
"type": "string",
"description": "Markdown: ## Goal / ## Boundary / ## Acceptance"
},
"files": {
"type": "array",
"items": {
"type": "string"
},
"description": "Implementing paths, ONE file per entry"
},
"derives_from": {
"type": "array",
"items": {
"type": "string"
},
"description": "Context note title(s) this system derives from → wires provenance (upstream + siblings)"
},
"batch": {
"type": "string",
"description": "import_from_code batch id"
},
"acknowledge_rejection": {
"type": "boolean",
"description": "Only after a DECLINED bounce AND asking the user"
},
"project_id": {
"type": "string"
}
},
"required": [
"title",
"spec"
]
}🟡propose_context(title, content, batch, resync, acknowledge_rejection, ...)
Propose a new/updated Idea note → Inbox. Title-match to update; send the COMPLETE revised text. Set resync:true ONLY when you rewrote the note FROM the current systems (get_stale lists notes the systems have moved past) — it stops the adopted note from immediately nagging to re-generate the systems it was just written from.
输入模式
{
"type": "object",
"properties": {
"title": {
"type": "string",
"description": "Match existing to update, or new to add"
},
"content": {
"type": "string",
"description": "Complete text, not a delta"
},
"batch": {
"type": "string",
"description": "import_from_code batch id"
},
"resync": {
"type": "boolean",
"description": "This note was re-derived from the current systems — re-stamps both staleness signals on adopt"
},
"acknowledge_rejection": {
"type": "boolean",
"description": "Only after a DECLINED bounce AND asking the user"
},
"project_id": {
"type": "string"
}
},
"required": [
"title",
"content"
]
}⚪propose_dna(dna, tech_notes, batch, project_id)
Propose Project DNA and/or Tech Notes → Inbox (at least one).
输入模式
{
"type": "object",
"properties": {
"dna": {
"type": "string",
"description": "Omit to leave unchanged"
},
"tech_notes": {
"type": "string",
"description": "Omit to leave unchanged"
},
"batch": {
"type": "string",
"description": "import_from_code batch id"
},
"project_id": {
"type": "string"
}
}
}🟡update_system(id, name, spec, new_title, context_title, ...)
Propose a system UPDATE → Inbox diff. get_system first; send the FULL revised spec. new_title renames; context_title+context bundles the Idea update. Set resync:true ONLY when you rewrote this spec FROM its upstream Idea (see resync_from_idea) — it re-stamps the staleness signal on adopt.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"spec": {
"type": "string",
"description": "Full revised spec markdown"
},
"new_title": {
"type": "string",
"description": "Rename in the same card"
},
"context_title": {
"type": "string",
"description": "Idea section to also update"
},
"context": {
"type": "string",
"description": "Full revised Idea text"
},
"resync": {
"type": "boolean",
"description": "This spec was re-derived from its upstream Idea — re-stamps sourceSig on adopt. Never set it on an ordinary edit."
},
"acknowledge_rejection": {
"type": "boolean",
"description": "Only after a DECLINED bounce AND asking the user"
},
"project_id": {
"type": "string"
}
},
"required": [
"spec"
]
}🟢get_balance(board, include_rows, project_id)
Read Balance Lab: stat tables, boards with EVALUATED values, scenarios.
输入模式
{
"type": "object",
"properties": {
"board": {
"type": "string",
"description": "One board (omit = all)"
},
"include_rows": {
"type": "boolean",
"description": "false = schema only"
},
"project_id": {
"type": "string"
}
}
}⚪propose_balance_table(title, csv, schema, rows, registry, ...)
Propose a stat-doc TABLE → Inbox. csv (header; `key%` = percent; first column = name) OR schema+rows. Adopting REPLACES same-named.
输入模式
{
"type": "object",
"properties": {
"title": {
"type": "string",
"description": "Table name"
},
"csv": {
"type": "string",
"description": "Header + rows; non-numeric = info"
},
"schema": {
"type": "array",
"items": {
"type": "object"
},
"description": "[{key, mode: flat|pct|text, base, min, max, desc}]"
},
"rows": {
"type": "array",
"items": {
"type": "object"
},
"description": "[{name, atk: 350, …}]"
},
"registry": {
"type": "boolean",
"description": "true = the shared Stats registry"
},
"project_id": {
"type": "string"
}
},
"required": [
"title"
]
}⚪propose_balance_board(name, nodes, project_id)
Propose a node board → Inbox. Kinds const|formula|item|sheet|loadout|picker|pool|process|note|frame; formulas reference blocks by NAME; omit x/y = auto-layout. Adopting REPLACES same-named.
输入模式
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Board name"
},
"nodes": {
"type": "array",
"items": {
"type": "object"
},
"description": "e.g. {kind:\"formula\",name:\"dmg\",formula:\"atk - mdef\"}"
},
"project_id": {
"type": "string"
}
},
"required": [
"name",
"nodes"
]
}⚪propose_milestone(name, goal, tasks, acknowledge_rejection, project_id)
Propose a NEW milestone (+ optional tasks) → Inbox.
输入模式
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Milestone name"
},
"goal": {
"type": "string",
"description": "One-line goal (optional)"
},
"tasks": {
"type": "array",
"items": {
"type": "string"
},
"description": "Task names (optional)"
},
"acknowledge_rejection": {
"type": "boolean",
"description": "Only after a DECLINED bounce AND asking the user"
},
"project_id": {
"type": "string"
}
},
"required": [
"name"
]
}⚪report_milestone_progress(id, milestone, done_tasks, project_id)
Mark existing-milestone tasks done → Inbox progress card (owner adopts the ticks).
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"milestone": {
"type": "string",
"description": "Milestone name (if no id)"
},
"done_tasks": {
"type": "array",
"items": {
"type": "string"
},
"description": "Task names that are done"
},
"project_id": {
"type": "string"
}
},
"required": [
"done_tasks"
]
}🟢report_build_status(id, name, status, files, decisions, ...)
Report a SYSTEM's status (todo|in-progress|implemented) — DIRECT, live. ALWAYS pass files (full list — it REPLACES); non-todo with no files shows done-but-EMPTY. 'implemented' counts as VERIFIED only when every UNIT-TESTABLE ## Acceptance bullet is backed via evidence[{criterion,proof}]; a bullet tagged [manual]/[e2e]/[ui]/[wip] is EXEMPT (verified by manual/e2e). Otherwise it's a CLAIM (verified:false) and the response names the unbacked criteria. TWO tiers: VERIFIED = a test is NAMED; GREEN (guarantee) = you RAN the test and reported `passed:true` on the evidence — green goes stale after 21 days, so re-run to keep it.
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "If no id"
},
"status": {
"type": "string",
"description": "todo | in-progress | implemented"
},
"files": {
"type": "array",
"items": {
"type": "string"
},
"description": "ONE file per entry (optional ' — function'); prose → decisions"
},
"decisions": {
"type": "array",
"items": {
"type": "string"
},
"description": "Decisions/divergences (prose)"
},
"evidence": {
"type": "array",
"items": {
"type": "object",
"properties": {
"criterion": {
"type": "string"
},
"proof": {
"type": "string"
},
"passed": {
"type": "boolean",
"description": "true = you RAN this test and it passed → counts toward TEST-GREEN"
},
"at": {
"type": "string",
"description": "ISO time of the pass (defaults to now); green expires after 21d"
}
}
},
"description": "Map ## Acceptance bullets to a test/file: [{criterion, proof, passed?, at?}]. Re-send full set (replaces). proof-only = backed (claimed); passed:true = test-green (guarantee)."
},
"version": {
"type": "string",
"description": "e.g. \"0.379.0\""
},
"project_id": {
"type": "string"
}
},
"required": [
"status"
]
}⚪propose_flow_edge(from, to, label, project_id)
SECONDARY (user-driven UI): propose a link between two existing screens → Inbox.
输入模式
{
"type": "object",
"properties": {
"from": {
"type": "string",
"description": "Source screen name"
},
"to": {
"type": "string",
"description": "Destination screen name"
},
"label": {
"type": "string",
"description": "Transition label (optional)"
},
"project_id": {
"type": "string"
}
},
"required": [
"from",
"to"
]
}🟢propose_element(screen, label, type, x, y, ...)
SECONDARY (user-driven UI): propose an element onto a screen → Inbox. ALWAYS pass x/y/w/h (px on the screen's `resolution` from get_screen); no coords = unplaced pile.
输入模式
{
"type": "object",
"properties": {
"screen": {
"type": "string",
"description": "Existing screen name"
},
"label": {
"type": "string",
"description": "Element label, e.g. 'HP bar'"
},
"type": {
"type": "string",
"description": "button/bar/panel/text (default box)"
},
"x": {
"type": "number",
"description": "Left px"
},
"y": {
"type": "number",
"description": "Top px"
},
"w": {
"type": "number",
"description": "Default 120"
},
"h": {
"type": "number",
"description": "Default 40"
},
"note": {
"type": "string"
},
"system": {
"type": "string",
"description": "System it serves"
},
"project_id": {
"type": "string"
}
},
"required": [
"screen",
"label"
]
}⚪propose_screen(name, purpose, parent, project_id)
SECONDARY (user-driven UI): propose a NEW screen → Inbox. parent = popup over that screen; purpose grounds AI suggestions. Whole UI? design_ui_from_systems first.
输入模式
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Screen name (Title Case, English)"
},
"purpose": {
"type": "string",
"description": "What this screen is for"
},
"parent": {
"type": "string",
"description": "Make it a POPUP over this screen"
},
"project_id": {
"type": "string"
}
},
"required": [
"name"
]
}⚪report_screen_status(screen, status, project_id)
Report a SCREEN's status (todo|in-progress|implemented) → Inbox.
输入模式
{
"type": "object",
"properties": {
"screen": {
"type": "string",
"description": "Screen name"
},
"status": {
"type": "string",
"description": "todo | in-progress | implemented"
},
"project_id": {
"type": "string"
}
},
"required": [
"screen",
"status"
]
}⚪rename_screen(name, new_name, project_id)
Rename a screen IN PLACE — DIRECT; links/elements follow (id-referenced).
输入模式
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Exact CURRENT name"
},
"new_name": {
"type": "string",
"description": "New name (must not collide)"
},
"project_id": {
"type": "string"
}
},
"required": [
"name",
"new_name"
]
}🟡update_milestone(id, name, goal, weeks, new_name, ...)
Edit milestone goal/weeks/difficulty/name — DIRECT. new_name renames (never delete+recreate).
输入模式
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Preferred"
},
"name": {
"type": "string",
"description": "Current name (if no id)"
},
"goal": {
"type": "string"
},
"weeks": {
"type": "number"
},
"new_name": {
"type": "string",
"description": "Rename"
},
"difficulty": {
"type": "string",
"enum": [
"easy",
"medium",
"hard",
""
],
"description": "\"\" clears"
},
"project_id": {
"type": "string"
}
}
}🟡add_task(milestone_id, milestone, name, done, status, ...)
Add a task to a milestone — DIRECT. assignee:"me" claims on creation.
输入模式
{
"type": "object",
"properties": {
"milestone_id": {
"type": "string",
"description": "Preferred"
},
"milestone": {
"type": "string",
"description": "If no id"
},
"name": {
"type": "string",
"description": "Task name"
},
"done": {
"type": "boolean"
},
"status": {
"type": "string",
"enum": [
"in-progress",
"blocked"
]
},
"effort": {
"type": "string",
"enum": [
"S",
"M",
"L"
]
},
"assignee": {
"type": "string",
"description": "\"me\" claims it"
},
"human": {
"type": "boolean",
"description": "Human-only (GTM/validation/people work) — next_task & get_briefing skip it"
},
"project_id": {
"type": "string"
}
},
"required": [
"name"
]
}⚪next_task(dry_run, project_id)
DISPATCHER (multi-agent): atomically pick + claim the next task to build — walks milestones in order, skips human-only tasks and any task whose systems share code files with a task another agent already holds or is actively touching, so parallel agents spread out instead of colliding. Returns the claimed task + systems, or why none is free. dry_run:true peeks without claiming.
输入模式
{
"type": "object",
"properties": {
"dry_run": {
"type": "boolean",
"description": "Peek without claiming"
},
"project_id": {
"type": "string"
}
}
}🟡update_task(task_id, name, done, status, effort, ...)
Edit a task or CLAIM it — DIRECT. assignee:"me"+status:"in-progress" claims (errors if held; force:true for stale); done:true or assignee:"" releases.
输入模式
{
"type": "object",
"properties": {
"task_id": {
"type": "string",
"description": "From get_milestone"
},
"name": {
"type": "string"
},
"done": {
"type": "boolean",
"description": "true releases the claim"
},
"status": {
"type": "string",
"enum": [
"in-progress",
"blocked",
""
],
"description": "\"\" clears"
},
"effort": {
"type": "string",
"enum": [
"S",
"M",
"L",
""
],
"description": "\"\" clears"
},
"assignee": {
"type": "string",
"description": "\"me\" claims; \"\" releases"
},
"force": {
"type": "boolean",
"description": "Take over a STALE claim only"
},
"human": {
"type": "boolean",
"description": "true = human-only (next_task/get_briefing skip it); false clears"
},
"project_id": {
"type": "string"
}
},
"required": [
"task_id"
]
}社区
证据