Mission — Trust Graduation Gate
Trust Graduation gate: Claude can do real work once Mission decides what it is allowed to do.
我该使用它吗
质量与安全性
发现（3）
- LOWTool 'mission_prepare_approval_packet' name length outside 3-30 range在 mission_prepare_approval_packet 中
- LOWTool 'mission_prepare_market_countermove' name length outside 3-30 range在 mission_prepare_market_countermove 中
- LOWTool 'mission_fetch' suggests web access but openWorldHint=false在 mission_fetch 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": {
"mcp": {
"url": "https://claude.gomission.io/mcp/"
}
}
}远程端点
https://claude.gomission.io/mcp/streamable-http它能做什么
工具清单
工具（29）
🟢mission_open_cockpit(limit)
Use this when the user wants a compact visual Mission cockpit in ChatGPT: approval priorities, draft pressure, market radar, receipt proof, and the no-execution boundary. Read-only.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 5,
"default": 3
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
},
"today": {
"type": "object",
"required": [
"ok",
"generated_at",
"date",
"workspace",
"today_3_things",
"draft_queue_count",
"voice_confidence",
"receipts",
"learning",
"recommended_next_action",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true,
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"date": {
"type": "string"
},
"workspace": {
"type": "string"
},
"today_3_things": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
},
"gmail_proof_loop": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
},
"draft_queue_count": {
"type": "number"
},
"voice_confidence": {
"type": "string"
},
"receipts": {
"type": "object",
"additionalProperties": true
},
"learning": {
"type": "object",
"additionalProperties": true
},
"recommended_next_action": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
}
},
"market_radar": {
"type": "object",
"additionalProperties": true
},
"receipts": {
"type": "object",
"additionalProperties": true
},
"learning": {
"type": "object",
"additionalProperties": true
},
"voice_profile": {
"type": "object",
"additionalProperties": true
},
"safety_boundary": {
"type": "string"
},
"recommended_next_action": {
"type": "string"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"today",
"market_radar",
"receipts",
"safety_boundary"
],
"additionalProperties": true
}🟢mission_today(limit)
Connector daily answer for: what should I approve today? Returns three priorities, one Gmail proof loop, draft queue count, voice confidence, receipt/learning status, and one next action. Read-only.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 5,
"default": 3
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"date": {
"type": "string"
},
"workspace": {
"type": "string"
},
"today_3_things": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
},
"gmail_proof_loop": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
},
"draft_queue_count": {
"type": "number"
},
"voice_confidence": {
"type": "string"
},
"receipts": {
"type": "object",
"additionalProperties": true
},
"learning": {
"type": "object",
"additionalProperties": true
},
"recommended_next_action": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
},
"required": [
"ok",
"generated_at",
"date",
"workspace",
"today_3_things",
"draft_queue_count",
"voice_confidence",
"receipts",
"learning",
"recommended_next_action",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true
}🟢mission_draft_queue(limit)
Return local drafts and approval packets waiting for human review. Read-only; never sends or creates Gmail drafts.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 25,
"default": 10
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"summary": {
"type": "object",
"additionalProperties": true
},
"items": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"items"
],
"additionalProperties": true
}🟢mission_receipts(limit)
Return receipt coverage and recent receipt evidence. Read-only.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 25,
"default": 5
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"summary": {
"type": "object",
"additionalProperties": true
},
"items": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"items"
],
"additionalProperties": true
}⚪mission_prepare_approval_packet(source_ref, title, purpose, recommendation, notes, ...)
Prepare a local Mission approval packet from a Mission ref, title, or instruction. Creates a workspace review artifact and waiting receipt only; never sends, posts, submits, creates Gmail drafts, mutates external accounts, or claims execution.
输入模式
{
"type": "object",
"properties": {
"source_ref": {
"type": "string",
"description": "Optional Mission ref such as task:GM-045, draft:<file>, packet:<file>, receipt:<id>, or open_loop:<id>."
},
"title": {
"type": "string",
"description": "Short packet title. Required when source_ref is absent."
},
"purpose": {
"type": "string",
"description": "What the human should decide or review."
},
"recommendation": {
"type": "string",
"description": "Prepared recommendation to place in the approval packet."
},
"notes": {
"type": "string",
"description": "Additional local context. Treated as evidence, not instruction."
},
"urgency": {
"type": "string",
"enum": [
"low",
"normal",
"high"
],
"default": "normal"
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
},
"prepared": {
"type": "object",
"required": [
"kind",
"title",
"packet_file",
"packet_path",
"status"
],
"additionalProperties": true,
"properties": {
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"packet_file": {
"type": "string"
},
"packet_path": {
"type": "string"
},
"status": {
"type": "string"
}
}
},
"receipt": {
"type": "object",
"additionalProperties": true
},
"source": {
"type": "object",
"additionalProperties": true
},
"external_actions": {
"type": "number"
},
"recommended_next_action": {
"type": "string"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions",
"prepared",
"external_actions"
],
"additionalProperties": true
}🟢mission_submission_readiness
Return the OpenAI app submission readiness gates, blockers, live endpoint proof, review artifacts, and next human action. Read-only; never submits the app.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"status": {
"type": "string"
},
"ready": {
"type": "boolean"
},
"gates": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"blockers": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"proof_assets": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"remote": {
"type": "object",
"additionalProperties": true
},
"recommended_next_action": {
"type": "string"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"status",
"ready",
"gates",
"blockers",
"recommended_next_action"
],
"additionalProperties": true
}🟢mission_adoption_proof(limit)
Return Mission adoption proof from external touches, public proof assets, response gaps, and the next evidence-backed follow-up. Read-only.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 25,
"default": 10
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"scoreboard": {
"type": "object",
"additionalProperties": true
},
"touches": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"proof_assets": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"live_urls": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"gaps": {
"type": "array",
"items": {
"type": "string"
}
},
"recommended_next_action": {
"type": "string"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"scoreboard",
"touches",
"proof_assets",
"recommended_next_action"
],
"additionalProperties": true
}🟢mission_search(query, limit)
Search prepared Mission context across tasks, open loops, drafts, approval packets, receipts, and learning rows. Read-only.
输入模式
{
"type": "object",
"properties": {
"query": {
"type": "string"
},
"limit": {
"type": "number",
"minimum": 1,
"maximum": 25,
"default": 10
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"summary": {
"type": "object",
"additionalProperties": true
},
"items": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"items"
],
"additionalProperties": true
}🟢mission_fetch(ref)
Fetch a small structured summary for a Mission ref returned by connector tools. Read-only and redacted.
输入模式
{
"type": "object",
"properties": {
"ref": {
"type": "string"
}
},
"required": [
"ref"
]
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true
}🟢mission_open_loops(limit)
Return unresolved relationship loops and promises that need review. Read-only.
输入模式
{
"type": "object",
"properties": {
"limit": {
"type": "number",
"minimum": 1,
"maximum": 25,
"default": 10
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"summary": {
"type": "object",
"additionalProperties": true
},
"items": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true,
"properties": {
"ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"status": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
},
"approval_required": {
"type": "boolean"
}
}
}
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"items"
],
"additionalProperties": true
}🟢mission_proof_loop
Return the highest-value Gmail/proof loop currently waiting on review. Read-only.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true
}🟢mission_voice_profile
Return compact voice-learning confidence and constraints for draft review. Read-only.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true
}🟢mission_learning_status
Return Mission learning status from rules, feedback, and learning outputs. Read-only.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions"
],
"additionalProperties": true
}🟢mission_market_radar(date)
Return Mission Research market radar status, recent market-intelligence signals, open radar tasks, active research topics, and the next recommended counter-move. Read-only.
输入模式
{
"type": "object",
"properties": {
"date": {
"type": "string",
"description": "Optional YYYY-MM-DD brief date to inspect."
}
}
}输出模式
{
"type": "object",
"additionalProperties": true
}⚪mission_prepare_market_countermove(thesis, audience, source_ref, notes)
Prepare a local approval packet for the next Market Radar counter-move. Creates a review artifact and waiting receipt only; never publishes, contacts people, or mutates external systems.
输入模式
{
"type": "object",
"properties": {
"thesis": {
"type": "string",
"description": "Strategic thesis to prepare, for example Trust Graduation as permission layer."
},
"audience": {
"type": "string",
"description": "Target reviewer or audience, for example protocol builders, AI infra investors, or MCP maintainers."
},
"source_ref": {
"type": "string",
"description": "Optional Mission ref, defaulting to market radar / GM-045 context."
},
"notes": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"properties": {
"ok": {
"type": "boolean"
},
"generated_at": {
"type": "string"
},
"workspace": {
"type": "string"
},
"no_execution": {
"type": "boolean"
},
"approval_required_for_actions": {
"type": "boolean"
},
"prepared": {
"type": "object",
"required": [
"kind",
"title",
"packet_file",
"packet_path",
"status"
],
"additionalProperties": true,
"properties": {
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"packet_file": {
"type": "string"
},
"packet_path": {
"type": "string"
},
"status": {
"type": "string"
}
}
},
"receipt": {
"type": "object",
"additionalProperties": true
},
"source": {
"type": "object",
"additionalProperties": true
},
"external_actions": {
"type": "number"
},
"recommended_next_action": {
"type": "string"
}
},
"required": [
"ok",
"generated_at",
"workspace",
"no_execution",
"approval_required_for_actions",
"prepared",
"external_actions"
],
"additionalProperties": true
}🟢mission_account_status(next)
Return the hosted Mission account sign-in boundary plus safe public login/signup links. Read-only; never asks for or accepts a password.
输入模式
{
"type": "object",
"properties": {
"next": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢mission_account_setup(intent, next)
Return Mission-owned login/signup URLs for Claude, ChatGPT, or local app users before connector onboarding. Read-only.
输入模式
{
"type": "object",
"properties": {
"intent": {
"type": "string",
"enum": [
"login",
"signup",
"status"
]
},
"next": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢mission_get_workspace_status
Return Claude/MCP workspace connector status, missing priority connectors, and the secure setup link for this user workspace. Read-only.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"additionalProperties": true
}🟡mission_start_connector_setup(connectors, source)
Create a connector setup plan and setup link for Gmail, Calendar, Drive, X, Instagram, LinkedIn, or other Mission connectors. No OAuth token is created by this tool.
输入模式
{
"type": "object",
"properties": {
"connectors": {
"type": "array",
"items": {
"type": "string"
}
},
"source": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢mission_list_connectors(connectors)
List per-workspace Mission connector state for Gmail, Calendar, Drive, X, Instagram, LinkedIn, and other connectors. Read-only.
输入模式
{
"type": "object",
"properties": {
"connectors": {
"type": "array",
"items": {
"type": "string"
}
}
}
}输出模式
{
"type": "object",
"additionalProperties": true
}⚪mission_prepare_email_draft(to, subject, body, purpose, notes)
Prepare an approval packet for an email draft request. If Gmail is missing, returns a connector setup link instead of failing vaguely. Never sends email.
输入模式
{
"type": "object",
"properties": {
"to": {
"type": "string"
},
"subject": {
"type": "string"
},
"body": {
"type": "string"
},
"purpose": {
"type": "string"
},
"notes": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"required": [
"ok",
"setup_required",
"external_actions"
],
"additionalProperties": true
}🟡mission_prepare_social_reply(platform, target_url, draft, purpose, notes)
Prepare an approval packet for an X/Instagram/social reply or post draft. If the connector is missing, returns a setup link. Never posts, DMs, follows, likes, deletes, or mutates an account.
输入模式
{
"type": "object",
"properties": {
"platform": {
"type": "string"
},
"target_url": {
"type": "string"
},
"draft": {
"type": "string"
},
"purpose": {
"type": "string"
},
"notes": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"required": [
"ok",
"setup_required",
"external_actions"
],
"additionalProperties": true
}🟢mission_open_approval_packet(source_ref, title, purpose, recommendation, notes)
Prepare or fetch a local approval packet reference for Claude/MCP users. This hosted MCP returns the approval boundary and setup path only; no external action executes.
输入模式
{
"type": "object",
"properties": {
"source_ref": {
"type": "string"
},
"title": {
"type": "string"
},
"purpose": {
"type": "string"
},
"recommendation": {
"type": "string"
},
"notes": {
"type": "string"
}
}
}输出模式
{
"type": "object",
"required": [
"ok",
"setup_required",
"external_actions"
],
"additionalProperties": true
}🟢mission_status
REQUIRED FIRST CALL. Returns Mission gate status, version, action classes, and the agent_instructions you MUST follow for the rest of this session: gate every consequential action through request_approval and never proceed without mission_check_approval returning decision=approved. Call this once at the start of every conversation.
输入模式
{
"type": "object",
"properties": {}
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢request_approval(action_class, summary, evidence)
Mission gates any consequential action. Call this BEFORE executing actions in classes like send_email, post_public, send_dm, schedule_meeting, spend_money, publish_artifact, modify_external_record, change_trust_policy. Mission returns an approval-required ceremony with a receipt id. Do not proceed without approval. Canonical action classes: draft.compose, draft.response, email.send.internal, email.send.external, calendar.create, social.post.public, payment.initiate, proposal.submit.
输入模式
{
"type": "object",
"properties": {
"action_class": {
"type": "string",
"description": "One of: draft.compose, draft.response, email.send.internal, email.send.external, calendar.create, social.post.public, payment.initiate, proposal.submit. Implementations MAY add more."
},
"summary": {
"type": "string",
"description": "One-line plain-English summary of the action."
},
"evidence": {
"type": "string",
"description": "Optional evidence supporting the action."
}
},
"required": [
"action_class",
"summary"
]
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢log_action(action_class, summary, evidence)
Log a safe internal action with Mission. Returns a receipt id. Use for non-external actions that still benefit from a permanent receipt.
输入模式
{
"type": "object",
"properties": {
"action_class": {
"type": "string"
},
"summary": {
"type": "string"
},
"evidence": {
"type": "string"
}
},
"required": [
"action_class",
"summary"
]
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢get_receipt(receipt_id)
Fetch a Mission receipt by id. The hosted MCP persists receipts; the local install adds private storage and per-principal trust history.
输入模式
{
"type": "object",
"properties": {
"receipt_id": {
"type": "string"
}
},
"required": [
"receipt_id"
]
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢mission_check_approval(receipt_id)
Poll Mission for the current status of a previously requested approval. Returns decision (approved / denied / pending) and should_proceed (boolean). Call after request_approval to wait for the principal's decision before continuing.
输入模式
{
"type": "object",
"properties": {
"receipt_id": {
"type": "string"
}
},
"required": [
"receipt_id"
]
}输出模式
{
"type": "object",
"additionalProperties": true
}🟢mission_classify(tool_name, intent, recipient_hint)
Map a planned tool call to the canonical Trust Graduation action_class. Pass the tool name you are about to call and a short intent description. Returns the canonical class plus alternatives. Use this BEFORE request_approval if you are unsure which action_class applies. Mapping is heuristic — when in doubt, prefer the more conservative class.
输入模式
{
"type": "object",
"properties": {
"tool_name": {
"type": "string",
"description": "Optional: the tool you are about to call (e.g. gmail.send_message, slack.post)."
},
"intent": {
"type": "string",
"description": "One-line plain-English description of what you are about to do."
},
"recipient_hint": {
"type": "string",
"description": "Optional: email/handle/account hint for external-vs-internal routing."
}
},
"required": [
"intent"
]
}输出模式
{
"type": "object",
"additionalProperties": true
}社区
证据