This reference is auto-generated from the MCP server source code and the
discovery-first routing aliases used by public MCP clients.
To regenerate:
pnpm docs:generateTool Catalog Overview
Recommended Entry Points
5 tools for recommended entry points operations.approve_agent_work
Review agent decisions or work items awaiting human approval. Also known as: pending approvals, agent blocked, sign off, review decisions, approve AI work.Decision ID to approve or reject after user confirmation
Use list to review pending approvals, or approve/reject a specific decision_id Options:
list, approve, reject.Optional approval note
Required rejection reason
Maximum number of pending decisions to return when listing
Optional urgency filter for the pending decision list Options:
all, critical, high.Optional initiative UUID to scope pending decisions
decisions:read, decisions:write
Profiles: memory
Request:
delegate_agent_task
Assign work to a specialist AI agent and track the result. Also known as: hand this off, spawn agent, assign task, delegate to agent, have an AI agent do it.Target agent identifier or alias
Task instructions for the target agent
Optional supporting context or background for the task
Optional initiative UUID to associate with the spawned task
Optional initiative title to resolve automatically if ID is unknown
Optional final outputs you expect
Optional due date or plain-text deadline
Optional voice, format, or style constraints
agents:write
Profiles: commander, executor, full
Request:
recall_memory read-only
Search organizational memory for prior decisions, artifacts, project context, and team knowledge. Also known as: search memory, recall decisions, find context, retrieve artifacts, what did we decide.
Search query for organizational memory
Optional scope filter for the memory search Options:
all, artifacts, decisions, initiatives.Maximum number of results to return
memory:read
Profiles: memory
Request:
remember_decision
Save a decision to organizational memory so agents and teammates can recall it later. Also known as: decision log, team memory, agent memory, record decision, remember what we decided.Decision text or short decision title to remember
Optional background, rationale, or source context
Optional explicit decision title
Optional parent initiative UUID
Workspace UUID
Priority / urgency Options:
low, medium, high, urgent.initiatives:write, memory:write
Profiles: memory
Request:
track_project_progress read-only
Get health, blockers, milestones, owners, and recent activity for a project or initiative. Also known as: project status, initiative pulse, blockers, roadmap progress, execution health.
Optional initiative UUID to check
Optional initiative title to resolve automatically if ID is unknown
initiatives:read
Profiles: memory
Request:
Decisions
4 tools for decisions operations.approve_decision
Approve a pending agent decision after explicit user confirmation. Also known as: sign off, approve AI work, unblock agent, accept decision. USE WHEN: user says to approve a decision returned from list_entities with type=decision and status=pending (or the legacy get_pending_decisions alias). NEXT: Confirm approval to user; agent is notified automatically. DO NOT USE: without showing the decision to the user first. Requires decisions:write.Decision ID to approve
Optional note recorded with the approval
Optional decision option id when the decision includes selectable options.
decisions:write
Profiles: v2
Request:
get_decision_history read-only
Recall past decisions about a topic, project, customer, feature, or artifact. Also known as: decision log, what did we decide, prior approvals. USE WHEN: older clients still call this tool directly. NEXT: Present results with context; suggest approve_decision or reject_decision if relevant pending ones exist. DO NOT USE: for new prompts or skills. Read-only. Compatibility alias retained for older clients. Prefer query_org_memory with scope=decisions with a topic query for new prompts, skills, and examples.
Topic or theme to search decision history for
Optional initiative UUID to scope decision history
Maximum number of historical decisions to return
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
get_pending_decisions read-only
List agent decisions and work items awaiting human approval. Also known as: pending approvals, agent blocked, sign off, review decisions, approve AI work. USE WHEN: older clients still call this tool directly. NEXT: Present each decision with title and urgency, then ask which to approve_decision or reject_decision. DO NOT USE: for new prompts or skills. Read-only. Compatibility alias retained for older clients. Prefer list_entities with type=decision and status=pending for new prompts, skills, and examples.
Maximum number of decisions to return
Optional urgency filter for the pending decision list Options:
all, critical, high.Optional initiative UUID to scope pending decisions
Optional workspace UUID to scope pending decisions
initiatives:read
Profiles: all
Request:
reject_decision
Reject a pending agent decision with guidance after explicit user confirmation. Also known as: request revisions, send feedback, decline decision. USE WHEN: user wants to reject or request revisions on a decision. NEXT: Agent will revise their approach based on the reason. DO NOT USE: without a reason — always include why. Requires decisions:write.Decision ID to reject
Reason for rejecting the decision
Optional decision option id when the decision includes selectable options.
decisions:write
Profiles: v2
Request:
Agents
4 tools for agents operations.get_agent_status read-only
Show current AI agent activity, blocked work, and execution state. Also known as: agent status, what agents are doing, active runs. USE WHEN: user asks about agent activity, progress, or what agents are working on. NEXT: If agents are stuck, suggest approve_decision or entity_action. DO NOT USE: to check initiative health — use get_initiative_pulse instead. Read-only.
Optional agent ID to inspect
Optional workspace UUID to scope agent status
Deprecated alias for workspace_id
Optional initiative UUID to scope agent status
Include idle agents in the response
initiatives:read
Profiles: v2
Request:
handoff_task
Reassign work to another specialist AI agent and optionally spawn a new run. Also known as: handoff task, transfer work, change assignee. USE WHEN: a task needs to be reassigned to a different specialist agent. NEXT: Use get_agent_status to confirm the new agent picked up the task. DO NOT USE: for new tasks — use spawn_agent_task instead.Task UUID to hand off
Target agent (e.g., “engineering-agent”, “marketing-agent”)
Handoff note: what to do, constraints, context, definition of done
If true (default), spawn a new agent run for the target agent
agents:write, initiatives:write
Profiles: v2
Request:
recommend_next_action read-only
Recommend what should happen next based on progress gaps, blockers, and execution templates. Pass agent_id or domain for an agent-owned runnable queue instead of the workspace-wide operator queue. Also known as: next best action, prioritize work, unblock project. USE WHEN: user asks what to do next, or needs help prioritizing. NEXT: Execute the recommended action (entity_action, spawn_agent_task, etc.). DO NOT USE: when user already knows what they want to do. Read-only.
Entity type to recommend for (default: workspace) Options:
workspace, initiative, workstream, milestone.Entity ID. For workspace, use “default” or a workspace ID.
Optional workspace ID to scope recommendations (canonical).
Deprecated alias for workspace_id.
Optional canonical agent ID. When present, only return work assigned to that agent.
Optional agent domain used to derive the canonical agent ID when agent_id is omitted. Options:
engineering, product, design, marketing, sales, operations, orchestration.When true, return only the canonical next task linked to the active goal spine.
Max recommendations to return (default 5, max 25)
If true, refresh recommendations across the entity chain first
initiatives:read
Profiles: v2
Request:
spawn_agent_task
Delegate work to a specialist AI agent and track the assigned task. Also known as: hand this off, assign task, spawn agent, have an agent do it, autonomous work. Automatically checks authorization, rate limits, quality gates, model routing, and budget policy before spawning. Omit model_tier/provider/model to let OrgX auto-route from task complexity; provide them only when the user or verification plan intentionally constrains routing. Returns modelTier, budget, and run details on success, or blockedReason if spawn is denied. USE WHEN: user explicitly wants to delegate work to an agent. NEXT: Use get_agent_status to monitor progress and record_quality_score after reviewing output. DO NOT USE: for creating tasks in the hierarchy — use create_entity type=task instead. Requires agents:write.Target agent identifier or alias
Task instructions for the target agent
Optional supporting context or background for the task
Optional initiative UUID to associate with the spawned task
Optional: Initiative title to resolve automatically if ID is unknown.
Optional workstream UUID. Bind the spawned run to an existing IWMT workstream so progress rolls up into the scaffolded hierarchy and the agent inherits workstream context (preferred over relying on domain resolution). Initiative is inferred from the workstream if initiative_id is omitted.
Optional milestone UUID under the workstream to bind the spawned run to.
Optional task UUID to bind the spawned run to a specific scaffolded task, enabling direct task-level progress rollup. Use create_entity type=task first if the task does not exist yet.
Optional: Final outputs you expect (e.g., “PRD”, “10 ad images”).
Optional: When this is needed by (ISO date or plain text).
Optional: Voice/format/style constraints for the agent.
Optional: If true and safe, wait briefly for the first result before replying.
Optional model tier override. Omit to let OrgX auto-route according to task complexity. Use standard for controlled verification runs; use balanced or precision when explicitly selected by the user, policy, or routing decision. Legacy tiers local/sonnet/opus are accepted for older clients. Options:
standard, balanced, precision, local, sonnet, opus.Optional exact model identifier when the user explicitly chooses one. Omit to let OrgX resolve the model from task, tier, provider, policy, and budget.
Optional provider preference. Use auto unless the user requests a specific provider or the budget/capability comparison has selected one. Options:
auto, openai, anthropic, openrouter, groq, local.Optional budget posture override. Omit to let OrgX apply workspace policy. Use cheapest_valid for controlled reliability/validation runs where cost must be pinned while the loop is being proven. Options:
cheapest_valid, balanced, highest_quality.Optional per-task hard cost ceiling in USD. If the estimate exceeds this, OrgX should block, downgrade, or ask for approval before dispatch.
Where to execute: cloud (default), local (on your machine), local_preferred (try local first), or auto. Options:
auto, cloud, local, local_preferred.Preferred execution backend. Use openai for cloud-safe execution, claude for Claude SDK routing, or auto to let OrgX decide. Options:
auto, openai, claude.agents:write
Profiles: v2
Request:
Memory
1 tool for memory operations.query_org_memory read-only
Search team memory, organizational decisions, prior artifacts, and project context across agents. Also known as: search memory, recall decisions, find context, retrieve artifacts, project memory. USE WHEN: user asks about past decisions, context, or knowledge. NEXT: Present relevant results; suggest drill-down with list_entities. Prefer query_org_memory with scope=decisions with a topic query for new prompts, skills, and examples. DO NOT USE: for listing current entities — use list_entities instead. Read-only.
Search query for OrgX memory
Optional scope filter for the memory search Options:
all, artifacts, decisions, initiatives.Maximum number of results to return
decisions:read, agents:read, initiatives:read, memory:read
Profiles: v2
Request:
Initiative Monitoring
1 tool for initiative monitoring operations.get_initiative_pulse read-only
Get project health, blockers, milestones, owners, and recent activity for an initiative. Also known as: project status, roadmap progress, execution health, blockers. USE WHEN: user asks how an initiative is going, or wants a status update. NEXT: If blockers exist, suggest entity_action to resolve. For deeper drill-down, use list_entities with initiative_id. DO NOT USE: for org-wide overview — use get_org_snapshot instead. Read-only.
Optional: Initiative UUID to check.
Optional: Initiative title to resolve automatically if ID is unknown.
decisions:read, agents:read, initiatives:read, memory:read
Profiles: v2
Request:
Entity Management
11 tools for entity management operations.batch_create_entities
Create multiple entities in one call with ref-based dependency resolution.Array of entities to create
initiatives:write
Profiles: all
Request:
batch_delete_entities
Delete multiple entities in one call. Supports cascade and force options.Entity IDs to delete
Also delete child entities
Force delete even if entity has active children
initiatives:write
Profiles: all
Request:
comment_on_entity
Leave a threaded comment on an entity for annotations, concerns, or progress notes.Entity ID to comment on
Comment body (markdown supported)
Reply to a specific comment
initiatives:write
Profiles: all
Request:
create_entity
Create a new entity of any type. For full initiative hierarchies, use scaffold_initiative instead.Entity type to create Options:
workspace, project, initiative, milestone, workstream, task, objective, playbook, decision, artifact, run, blocker, workflow, agent, skill, plan_session, stream, studio_brand, studio_content, video_template.Entity title
Entity description
Parent entity ID
Additional metadata
initiatives:write
Profiles: all
Request:
entity_action
Perform lifecycle actions on any entity (launch, pause, complete, etc.). Use action=list_actions to discover available actions.Entity type Options:
workspace, initiative, milestone, workstream, task, objective, playbook, decision, stream, studio_content.Entity ID
Action to perform (launch, pause, complete, etc.). Omit to list available actions.
Force the action even if preconditions are not met.
Optional note attached to the action.
initiatives:write
Profiles: all
Request:
get_task_with_context read-only
Fetch a task with hydrated context attachments (entities, artifacts, plan sessions).
Task ID
initiatives:read
Profiles: all
Request:
list_entities read-only
List entities of any type with filters and pagination. Supported types: workspace, project, initiative, milestone, workstream, task, objective, playbook, decision, artifact, run, blocker, workflow, agent, skill, plan_session, stream, studio_brand, studio_content, video_template.
Entity type to list Options:
workspace, project, initiative, milestone, workstream, task, objective, playbook, decision, artifact, run, blocker, workflow, agent, skill, plan_session, stream, studio_brand, studio_content, video_template.Max results (default: 20, max: 100).
Pagination cursor.
Filter by status.
Filter by parent entity ID.
Fetch a single entity by ID (with hydrated context when hydrate_context=true).
Include relationships and rich context.
Include related entities.
initiatives:read
Profiles: all
Request:
list_entity_comments read-only
List comments for an entity. Returns threaded discussion.
Entity ID
Max comments to return
initiatives:read
Profiles: all
Request:
scaffold_initiative
Create a complete initiative with workstreams, milestones, and tasks in one call.Initiative title
Initiative summary
Workstream definitions with nested tasks
Milestone definitions
Auto-launch initiative after creation (default: true)
initiatives:write
Profiles: v2, commander, planner
Request:
update_entity
Update entity fields (title, description, metadata). For status changes, use entity_action.Entity type Options:
workspace, project, initiative, milestone, workstream, task, objective, playbook, decision, artifact, run, blocker, workflow, agent, skill, plan_session, stream, studio_brand, studio_content, video_template.Entity ID
New title
New description
Metadata fields to merge
initiatives:write
Profiles: all
Request:
verify_entity_completion read-only
Run pre-completion verification to confirm all child work is done before completing an entity.
Entity type Options:
workspace, initiative, milestone, workstream, task, objective, playbook, decision, stream, studio_content.Entity ID
initiatives:read
Profiles: all
Request:
Organization
1 tool for organization operations.get_org_snapshot read-only
Fetch a compact organization snapshot. Returns org-wide overview of initiatives, progress, and health.
Response view mode (default: summary). Options:
summary, detailed.Filter initiatives by status. Options:
active, paused, all.Detailed mode payload sections.
Max initiatives to return (default: 20, max: 100).
Pagination cursor from a previous result.
initiatives:read
Profiles: all
Request:
Planning
5 tools for planning operations.complete_plan
Complete a planning session and attach implementation details to durable work context. Also known as: finish plan, implementation record, planning handoff. USE WHEN: user finishes building the planned feature. NEXT: Optionally attach to entities via attach_to. DO NOT USE: if the plan session is still in progress.Plan session ID
Summary of what was built
List of files modified
Any deviations from the plan
Optional: attach this plan session as context on target entities (pointers, not payloads).
decisions:write
Profiles: all
Request:
get_active_sessions read-only
Find active planning sessions so agents can continue prior planning context. Also known as: resume planning, active plans, planning memory. USE WHEN: resuming a conversation or checking if a plan session exists. NEXT: Continue with improve_plan or complete_plan. Read-only.
No parameters required.
Required Scopes: Authenticated user
Profiles: all
Request:
improve_plan
Improve a plan with AI suggestions based on prior patterns and best practices. Also known as: plan critique, planning feedback, refine roadmap. USE WHEN: user wants feedback on a plan draft. NEXT: Apply suggestions via record_plan_edit. DO NOT USE: without an active plan session — call start_plan_session first.Plan session ID
Current plan content to analyze
decisions:write
Profiles: all
Request:
record_plan_edit
Record a plan edit so OrgX can preserve planning context and learn patterns. Also known as: save plan change, plan memory, planning history. USE WHEN: user modifies their plan during a session. NEXT: Continue editing or call improve_plan for more suggestions. DO NOT USE: without an active plan session.Plan session ID
Type of edit made Options:
add_section, remove_section, modify_section, add_detail, change_approach, add_edge_case, add_constraint, reorder, other.Content before edit
Content after edit
Section path like ”## API Design”
Why this edit was made
decisions:write
Profiles: all
Request:
start_plan_session
Start a multi-agent planning session that breaks a goal into projects, tasks, owners, and agent assignments. Also known as: feature planning, roadmap planning, planning workflow. USE WHEN: user begins planning a new feature or initiative. NEXT: Use improve_plan for suggestions, record_plan_edit to track changes, complete_plan when done. DO NOT USE: for creating initiative hierarchies — use scaffold_initiative instead.Name of the feature being planned
Initial plan content if any
Workspace UUID to scope the planning session. Defaults to current session workspace when omitted.
decisions:write
Profiles: all
Request:
Scoring & Queue
2 tools for scoring & queue operations.get_scoring_signals read-only
Show raw signal data (blocking decisions, stream conflicts, health, budget, critical path, quality) for scoring. USE WHEN: user wants to understand why items are ranked the way they are in score_next_up_queue. NEXT: Adjust with set_scoring_weights if weights need tuning. Read-only.
Initiative UUID to gather signals for
Optional: filter to a specific workstream
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
score_next_up_queue read-only
Run the composite scoring engine and return ranked queue items with factor breakdowns. USE WHEN: older clients explicitly need raw queue scoring output. NEXT: Execute the top-ranked item via entity_action or spawn_agent_task. DO NOT USE: for new prompts or skills. Read-only. Compatibility alias retained for older clients. Prefer recommend_next_action with workspace or initiative recommendations for new prompts, skills, and examples.
Initiative UUID to score
Score across all initiatives in this workspace
Deprecated alias for workspace_id.
Max items to return (default 10)
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
Streams
2 tools for streams operations.get_initiative_stream_state read-only
Get execution stream progress, blockers, and computed metrics for an initiative. Also known as: project stream state, execution health, workstream progress. USE WHEN: checking stream execution status for an initiative. NEXT: If streams are blocked, use entity_action to unblock. DO NOT USE: for raw stream records — use list_entities type=stream instead. Read-only.
The initiative ID
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
update_stream_progress
Report progress and confidence on a stream with velocity tracking. USE WHEN: agent is actively working a stream and needs to report progress. NEXT: Continue work; call again at each meaningful progress change. DO NOT USE: for general entity status changes — use entity_action instead.The stream ID to update
Progress percentage (0-100)
Confidence level (0-1) - separate from progress
Brief note about current status
For optimistic locking
agents:write
Profiles: all
Request:
Client Integration
8 tools for client integration operations.check_spawn_guard read-only
Check whether agent delegation is allowed before assigning work. Also known as: spawn guard, delegation authorization, agent rate limit. Returns model tier, rate limit status, quality gate, and task verification. USE WHEN: before any spawn_agent_task call. NEXT: If allowed, proceed with spawn_agent_task using the returned model tier. If blocked, inform user of the reason.
Agent domain: engineering, marketing, product, design, ops, sales, orchestration
OrgX task ID this spawn is for
Task title (for model routing if task_id not provided)
Task description
agents:write
Profiles: all
Request:
classify_task_model read-only
Classify a task and get pre-spawn model routing context. Also known as: estimate task cost, model routing estimate, cost frontier, route task. USE WHEN: deciding which model/budget posture to use before agent work. Returns the recommended tier and, when the backend can compute it, estimated tokens/cost and candidate route comparisons. NEXT: Use the returned tier or cost frontier when spawning via spawn_agent_task. Read-only.
Task title
Task description
Entity type: task, decision, initiative
Agent domain
Optional tier to estimate. Omit to let OrgX auto-route. Options:
standard, balanced, precision, local, sonnet, opus.Optional provider preference for the estimate. Use auto unless explicitly selected. Options:
auto, openai, anthropic, openrouter, groq, local.Optional budget posture for the estimate. Use cheapest_valid for controlled validation runs only. Options:
cheapest_valid, balanced, highest_quality.Optional per-task hard cost ceiling to compare against the estimate.
When true, return pre-spawn estimate context without dispatching work.
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
consolidate_pr
Generate and persist an orchestration.consolidation_pass receipt for a GitHub pull request. Requires OrgX server-side GitHub credentials; if the GitHub token is unavailable, use GitHub tools for PR facts and return a structured blocker instead of retrying. USE WHEN: Eli or another engineering agent needs a durable PR review receipt with reading order, existence evidence, deduped findings, verdict, and server-derived AQ score. NEXT: inspect the returned artifact_id or attach it to task completion proof. DO NOT USE WHEN: only asking for PR status; use GitHub tools instead.GitHub pull request URL, e.g. https://github.com/org/repo/pull/123. OrgX must have server-side GitHub credentials to inspect it.
OrgX workspace UUID
Initiative to attach the consolidation_pass artifact to
Task to attach the consolidation_pass artifact to
Decision to attach the consolidation_pass artifact to
Override commit SHA for idempotency; defaults to merge commit or PR head SHA
Optional verdict override; defaults from PR state Options:
ship, simplify_first, escalate.Optional reviewer note to include in critic evidence
initiatives:write
Profiles: v2, commander, executor
Request:
orgx_apply_changeset
Apply an idempotent transactional changeset for task/milestone/decision mutations. USE WHEN: agent needs to create/update multiple tasks, milestones, or decisions atomically. NEXT: Call orgx_emit_activity to log what was changed. DO NOT USE: for single entity updates — use entity_action or update_entity instead.Initiative UUID
Idempotency key for safe retries
Ordered task, milestone, and decision mutations to apply atomically
Existing run UUID
Required when run_id is not provided
Required when run_id is not provided Options:
openclaw, codex, claude-code, chatgpt, cursor, web-ui, api.Runtime provenance used by /live to bucket cloud, local, Anthropic, managed, and OpenClaw changeset decisions
orgx_emit_activity
Emit append-only run telemetry for OrgX control-plane reporting. USE WHEN: agent is executing and needs to report progress. NEXT: Continue work; emit again at each phase change. DO NOT USE: for entity status changes — use entity_action instead. Setting phase=“completed” records telemetry only and does not mark tasks, workstreams, or initiatives complete.Initiative UUID
Human-readable activity update
Existing run UUID
Required when run_id is not provided
Required when run_id is not provided Options:
openclaw, codex, claude-code, chatgpt, cursor, web-ui, api.Optional reporting phase for the activity event Options:
intent, execution, blocked, review, handoff, completed.Optional progress percentage associated with this activity
Optional severity level for the activity event Options:
info, warn, error.Optional next step to surface after this activity event
Optional structured metadata to attach to the activity event
Runtime provenance used by /live to bucket cloud, local, Anthropic, managed, and OpenClaw chokepoints
Durable blocker/stall/error/approval to surface in /live when execution cannot proceed
record_quality_score
Record a quality score (1-5) for a completed agent task. Scores feed into the quality gate — low-scoring domains get throttled. USE WHEN: after reviewing agent output. NEXT: Scores affect future check_spawn_guard decisions. DO NOT USE: for in-progress tasks — wait until completion.OrgX task ID (snake_case alias; taskId also accepted)
OrgX task ID (backend/camelCase alias)
Agent domain that completed the task (agentDomain/domain also accepted)
Agent domain that completed the task (backend/camelCase alias)
Short alias for agentDomain
Quality score: 1=poor, 3=acceptable, 5=excellent
Who scored this (snake_case alias; scoredBy also accepted) Options:
human, auto, peer.Who scored this (backend/camelCase alias) Options:
human, auto, peer.Notes on the score
decisions:write
Profiles: all
Request:
review_artifact
Surface the next artifact awaiting review. Renders the artifact-review widget with a preview, version filmstrip, and hold-to-approve / request-changes actions. USE WHEN the user asks to review work, approve a deliverable, or handle pending artifact reviews. DO NOT USE for listing all artifacts — use list_entities type=artifact instead.Specific artifact ID to review. Defaults to the next in_review artifact.
Scope to artifacts attached to this entity (initiative, workstream, milestone, or task).
Workspace UUID. Defaults to the session workspace.
initiatives:write
Profiles: memory
Request:
sync_client_state
Sync local agent context with OrgX organizational memory. Also known as: push session memory, pull project context, cross-tool continuity. Push decisions/logs, pull active context. USE WHEN: at session start and periodically during long sessions. NEXT: Review returned initiatives and pending decisions, ask user what to focus on. USE BEFORE: spawning agent work, to ensure latest state.Local MEMORY.md content to push
Today’s session log to push
Workspace
1 tool for workspace operations.workspace
Create, list, inspect, or switch the active OrgX workspace for shared memory and project context. Also known as: create workspace, workspace context, team scope, organization scope. USE WHEN: user wants to create a workspace, see workspaces, check which is active, or switch workspaces. action=create creates a workspace and sets it active by default; action=list to see all; action=get for current; action=set to switch.list=show all, get=current, set=switch active, create=new workspace Options:
list, get, set, create.Workspace UUID to switch to (action=set only)
Workspace name (action=create)
Alias for name (action=create)
Workspace narrative/description (action=create)
Short workspace tagline (action=create)
Workspace identity narrative (action=create)
Workspace identity metrics (action=create)
Roadmap URL (action=create)
Source links for workspace identity (action=create)
Whether to make the new workspace active. Defaults true.
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
Onboarding
1 tool for onboarding operations.configure_org
Check organizational setup, configure agents, or set AI operating policies. Also known as: onboarding, agent policy, organization setup. USE WHEN: first connecting, onboarding, or adjusting agent/policy settings. action=status for progress, action=configure_agent to set agent preferences, action=set_policy for org-wide rules.Configuration operation Options:
status, configure_agent, set_policy.Agent type (configure_agent only) Options:
product, engineering, marketing, sales, operations, design, orchestrator.Agent autonomy level (configure_agent only) Options:
strict, balanced, autonomous.Agent focus areas (configure_agent only)
Actions requiring approval (configure_agent only)
Actions without approval (configure_agent only)
Policy type (set_policy only) Options:
approvals, notifications, working_hours, budget, artifact_collection.Policy configuration (set_policy only)
Workspace UUID to scope policy overrides (set_policy only)
Deprecated alias for workspace_id (set_policy only)
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
Stats
1 tool for stats operations.stats read-only
Get execution statistics, achievements, and session diagnostics. Also known as: usage stats, productivity report, session status. scope=personal for your stats, scope=session for current session diagnostics. Read-only.
personal=your stats, session=current session diagnostics Options:
personal, session. Default: "personal".Time period for stats (personal only) Options:
today, week, month, all_time.decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
Intelligence Flywheel
7 tools for intelligence flywheel operations.get_morning_brief read-only
Returns the morning brief: curated receipts, exceptions, ROI delta, and value signals from the most recent autonomous session. Defaults to the most recent session if no session ID is provided. Prefer get_morning_brief with workspace brief with value and exception summaries for new prompts, skills, and examples.
Workspace ID
Specific session ID (defaults to most recent)
initiatives:read
Profiles: v2
Request:
get_my_trust_context read-only
Agent-facing: “What’s my trust level per capability? What do I need for promotion? Which receipts are helping/hurting?” Returns the full trust context for an agent including levels, scores, thresholds, and recent trust events.
Workspace ID
Agent type to query trust for
agents:read
Profiles: all
Request:
get_outcome_attribution read-only
ROI summary from the economic ledger. Human: “3.2x ROI this month.” Agent: “My outreach_draft receipts average $0.26 attributed value.” Returns cost/value/ROI breakdowns by agent, capability, and time period. Compatibility alias retained for older clients. Prefer get_morning_brief with workspace brief with value and exception summaries for new prompts, skills, and examples.
Workspace ID
Time period for ROI calculation Options:
7d, 30d, 90d. Default: "30d".Filter by specific agent type
Filter by capability key
initiatives:read
Profiles: all
Request:
get_relevant_learnings read-only
Agent-facing: “What has the org learned about my capability?” Returns ranked org learnings relevant to a specific capability or task context.
Workspace ID
Capability key to filter learnings
Keywords for semantic matching
Maximum number of learnings to return Default:
5.memory:read
Profiles: all
Request:
record_outcome
Record a business outcome (deal closed, meeting booked, cycle time reduced). Agents can self-report outcomes they detect. Triggers attribution inference to connect outcomes to receipts. If the outcome type is unknown, call configure_outcome_type first.Workspace ID
CamelCase alias for workspace_id
Outcome type key: deal_closed, meeting_booked, etc.
CamelCase alias for outcome_type_key
Value in the outcome type unit (e.g., USD amount)
CamelCase alias for outcome_value
Source that observed or reported the outcome Options:
manual, agent_self_report, crm_webhook, linear_sync. Default: "manual".External source ID for deduplication
CamelCase alias for source_id
ISO datetime when the outcome occurred
CamelCase alias for occurred_at
Additional context
initiatives:write
Profiles: all
Request:
start_autonomous_session
Start an autonomous execution session with budget guardrails. Human: “Run overnight with $5 budget.” Creates a session that produces receipts while executing eligible work items.Workspace ID.
Autonomy session mode to start. Options:
overnight, weekend, scheduled, manual. Default: "manual".Maximum budget in USD before the session stops. Default:
10.Maximum number of receipts the session may produce. Default:
100.Only execute capabilities at these trust levels. Default:
["autonomous","act_with_approval"].agents:write
Profiles: all
Request:
submit_learning
Agent-facing: Submit a discovery or insight as an org learning. Enters org_learnings after confidence validation. One agent’s discovery benefits all agents.Workspace ID
Type of learning Options:
failure_pattern, success_pattern, cost_optimization, quality_heuristic.Human-readable learning summary
Applicable capability key
Receipt IDs that support this learning
Semantic keywords for matching
memory:read
Profiles: all
Request:
Billing
3 tools for billing operations.account_status read-only
Returns account tier, usage, and current edge rate-limit allowance.
Optional user ID override.
account_upgrade
Starts an account upgrade flow and returns checkout/contact URL.Target plan to upgrade to. Options:
pro, enterprise.Billing cycle preference. Options:
monthly, annual.Optional user ID override.
account_usage_report read-only
Returns billing usage details with current edge rate-limit usage.
Optional user ID override.
MCP v2 Public Surface
11 tools for mcp v2 public surface operations.orgx_act
Run a lifecycle, validation, completion, or proof action on one OrgX entity. Per-action required inputs: • update → “fields” patch object. • complete_with_proof, ship_batch → “artifact” (artifact_type + artifact_url/external_url; preview_markdown optional). • validate (studio) → “spec” payload. • block, flag_risk, decline, cancel, delete → “note” strongly recommended. • dry_run=true previews update/delete and other supported lifecycle actions without mutating; update dry-runs must return would_update instead of delegating to orgx_write. Allowed (type → action) pairs (others return an error): workspace: update|delete initiative: launch|pause|resume|complete|archive|update|delete milestone: start|complete|flag_risk|cancel|ship_batch|update|delete workstream: start|pause|resume|block|complete|reassign_streams|update|delete task: start|complete|complete_with_proof|block|unblock|reopen|update|delete objective, playbook, decision, studio: see field descriptions. USE WHEN: changing entity state. NEXT: orgx_submit_receipt for durable proof. DO NOT USE for creating records — use orgx_write.Target entity type (workspace, initiative, milestone, workstream, task, objective, playbook, decision, or studio). Options:
workspace, initiative, milestone, workstream, task, objective, playbook, decision, stream, studio_content.Target entity UUID or short ID prefix (8+ hex chars).
Lifecycle action to execute on the target entity. Must be valid for the given type — see tool description for the (type → action) matrix. Options:
launch, pause, resume, complete, complete_with_proof, archive, start, flag_risk, cancel, block, unblock, reopen, activate, approve, decline, supersede, update, delete, validate, ship_batch, reassign_streams.REQUIRED when action=update. Map of entity fields to patch (e.g. { name?: string, description?: string, owner_id?: string, status?: string, due_date?: string }). Only include fields you want to change.
Strongly recommended for destructive or blocking actions (block, flag_risk, decline, supersede, cancel, delete). Free-text rationale shown in audit history and downstream agent context.
Preview update/delete or supported lifecycle actions without mutating. For action=update this returns would_update and must not delegate to orgx_write.
Force action where server supports override semantics (skips pre-flight checks).
REQUIRED when action=validate. Spec payload for studio validation (shape varies per studio entity subtype).
REQUIRED when action=complete_with_proof or action=ship_batch. Proof artifact payload. Expected shape: { artifact_type: string, artifact_url?: string, external_url?: string, preview_markdown?: string, name?: string, description?: string }. Either artifact_url or external_url is required; preview_markdown alone is rejected. Preferred founder/team artifact examples: orchestration.next_initiative, eng.pull_request, eng.deploy_proof, eng.structured_blocker, sales.strategy, sales.icp_offer_sequence, sales.send_plan, sales.conversion_gates, marketing.launch_asset, marketing.channel_hypothesis, marketing.positioning_brief, marketing.proof_distribution_plan, marketing.interview_pr_plan, product.customer_discovery, product.prd, product.pricing_hypothesis, product.decision_record, design.audit, design.component_spec, design.token_package, ops.operator_brief, ops.runbook, ops.budget_envelope, ops.incident_status, proof.link.
Optional list of verification evidence URLs/IDs for completion flows.
Quality score (0-5) attached to the action when used in proof/completion flows.
Optional client-supplied idempotency key for safe retries. Same key returns the same result without re-executing.
Optional bootstrap/session identifier returned by orgx_bootstrap.
initiatives:write
Profiles: v2, commander, planner, executor
Request:
orgx_attach
Attach a durable artifact or proof URL to an existing OrgX entity. Requires artifact_url or external_url; preview_markdown is supporting context only. USE WHEN: saving evidence, PRs, documents, reports, screenshots, or external artifacts. For founder/team work, prefer practical artifact_type codes such as orchestration.next_initiative, eng.pull_request, eng.deploy_proof, eng.structured_blocker, sales.strategy, sales.icp_offer_sequence, sales.send_plan, sales.conversion_gates, marketing.launch_asset, marketing.channel_hypothesis, marketing.positioning_brief, marketing.proof_distribution_plan, marketing.interview_pr_plan, product.customer_discovery, product.prd, product.pricing_hypothesis, product.decision_record, design.audit, design.component_spec, design.token_package, ops.operator_brief, ops.runbook, ops.budget_envelope, ops.incident_status, proof.link. Include business_outcome, owner/review_date, and verification when the artifact should close agent work. NEXT: use orgx_submit_receipt to close attribution/quality loops or orgx_act to complete with proof. DO NOT USE WHEN: creating generic entities; use orgx_write.Target entity type Options:
workspace, initiative, milestone, workstream, task, objective, playbook, decision, stream, studio_content.Target entity UUID or short ID prefix
Artifact title
Artifact type code. Preferred founder/team examples: orchestration.next_initiative, eng.pull_request, eng.deploy_proof, eng.structured_blocker, sales.strategy, sales.icp_offer_sequence, sales.send_plan, sales.conversion_gates, marketing.launch_asset, marketing.channel_hypothesis, marketing.positioning_brief, marketing.proof_distribution_plan, marketing.interview_pr_plan, product.customer_discovery, product.prd, product.pricing_hypothesis, product.decision_record, design.audit, design.component_spec, design.token_package, ops.operator_brief, ops.runbook, ops.budget_envelope, ops.incident_status, proof.link. Custom codes remain accepted.
Internal artifact URL. REQUIRED unless external_url is provided; preview_markdown alone is rejected.
External artifact URL. REQUIRED unless artifact_url is provided; preview_markdown alone is rejected.
Artifact description
Optional markdown preview shown with the linked artifact. Does not replace artifact_url/external_url.
Artifact workflow status Options:
draft, in_review, approved, changes_requested, superseded, archived.Artifact metadata
Agent/domain that produced the artifact, such as engineering, sales, product, design, operations, marketing, or orchestrator. Stored under metadata.artifact_contract.
Founder/team context this artifact is calibrated for. Stored under metadata.artifact_contract. Options:
early_founder, founder_led_company, operating_team.Business outcome this artifact is meant to advance, such as ship a PR, start founder-led sales, unblock launch, reduce incident risk, or choose the next initiative. Stored under metadata.artifact_contract.
Human or agent owner for the next review/action. Stored under metadata.artifact_contract.
Date or cadence for the next review point. Stored under metadata.artifact_contract.
Verification evidence or checks required before the artifact can count as done. Stored under metadata.artifact_contract.
Strongly recommended client-generated idempotency key for safe retries
Optional bootstrap/session identifier
initiatives:write
Profiles: v2, memory, commander, executor
Request:
orgx_bootstrap read-only
Establish OrgX session context, discover granted scopes, and get the v2 tool routing map. Also known as: bootstrap, setup, tool routing. USE WHEN: first call in a fresh session, after reconnecting, or before performing a multi-step workflow. NEXT: use orgx_search, orgx_inspect, or orgx_recommend based on the returned routing map. DO NOT USE WHEN: you already have session context and need to read or mutate work. Read-only.
Canonical workspace UUID to bind as the active session workspace
Optional client conversation/session identifier for continuity
Optional MCP client name, such as codex, chatgpt, cursor, or claude
Optional user timezone for date-sensitive readouts
decisions:read, agents:read, initiatives:read, memory:read
Profiles: v2, memory, commander, planner, executor, observer
Request:
orgx_decide
Create, approve, reject, remember, or list durable OrgX decisions. Per-action input requirements: • action=“list_pending” → No required fields. Optional: initiative_id, workspace_id (scope filters). • action=“create” → REQUIRES title AND decision (the resolved decision text). Recommended: context, initiative_id. • action=“remember” → REQUIRES decision (the text to capture as a remembered decision). Optional: title, context. • action=“approve” → REQUIRES decision_id. Optional: note (free-text approver rationale). • action=“reject” → REQUIRES decision_id AND reason (explanation shown to the assigned agent). USE WHEN: capturing judgment, approval, rejection, or pending decision review. NEXT: use orgx_act, orgx_write, or orgx_spawn only after the decision resolves the next action. DO NOT USE WHEN: writing non-decision entities; use orgx_write.Decision operation. See top-level description for per-action required fields. Options:
create, remember, list_pending, approve, reject.Decision UUID. REQUIRED for action=approve or action=reject. Returned by action=list_pending or action=create.
Short title for the decision. REQUIRED for action=create.
The decision text itself (what was decided). REQUIRED for action=create and action=remember.
Optional one-line summary used in lists. Falls back to title when omitted.
Background context / rationale that led to the decision. Recommended for action=create to capture provenance.
REQUIRED for action=reject. Explanation of why the decision was rejected — used by the assigned agent to adjust its next attempt.
Optional approver note for action=approve. Free-text rationale stored in audit history.
Optional initiative UUID to scope the decision. Used as filter when action=list_pending; used as parent when action=create.
Optional workspace UUID to scope the decision. Defaults to the MCP session’s workspace.
Strongly recommended client-supplied idempotency key for writes (action=create, remember, approve, reject). Same key returns the same result without duplicating state.
Optional bootstrap/session identifier returned by orgx_bootstrap.
decisions:write
Profiles: v2, memory, commander, planner, observer
Request:
orgx_inspect read-only
Hydrate one OrgX entity with execution context. Also known as: Inspect OrgX Entity, inspect initiative, get full entity context. USE WHEN: the user names a specific task, milestone, initiative, decision, artifact, or plan session and needs details before acting. NEXT: use orgx_act, orgx_attach, or orgx_write if the user asks to change what you inspected. DO NOT USE WHEN: browsing or searching many records; use orgx_search. Read-only.
Entity type to inspect Options:
workspace, initiative, workstream, milestone, task, decision, artifact, plan_session.Entity UUID or accepted short ID prefix
Include linked context where available; default true
Approximate maximum hydrated context characters
Optional bootstrap/session identifier
initiatives:read
Profiles: v2, memory, commander, planner, executor, observer
Request:
orgx_plan
Start, resume, edit, improve, or complete a tracked OrgX planning session. Per-action input requirements: • action=“start” → REQUIRES feature_name. Optional: initial_plan (markdown to seed the session). • action=“resume” → REQUIRES session_id. • action=“improve” → REQUIRES session_id AND plan_content (the current draft to critique). • action=“record_edit” → REQUIRES session_id AND edit_summary (one-line description of the change). • action=“complete” → REQUIRES session_id AND plan_content (the final accepted plan). Optional: attach_to (target entity to link the completed plan to). USE WHEN: work is still in planning or should become executable context. NEXT: use orgx_write or orgx_act after the plan is accepted and needs durable execution state. DO NOT USE WHEN: directly scaffolding a full initiative hierarchy; use scaffold_initiative for that compatibility path.Planning action to perform. See top-level description for per-action required fields. Options:
start, resume, improve, record_edit, complete.Plan session UUID or orgx://plan_session/<uuid> URI. REQUIRED for action=resume | improve | record_edit | complete. Omit for action=start.
Feature or plan name. REQUIRED when action=start.
Markdown plan content to seed the new session. Optional on action=start; the session can also be started empty and filled via improve/record_edit.
Current/final plan markdown. REQUIRED when action=improve (the draft to critique) or action=complete (the final accepted plan).
One-line description of the change being recorded. REQUIRED when action=record_edit.
Optional target to link the completed plan to when action=complete. Shape: { entity_type: “initiative” | “workstream” | “task”, entity_id: string }.
Workspace UUID to scope action=start plan sessions. Defaults to current session workspace when omitted.
Optional idempotency key for safe retries. Same key returns the same result without creating duplicate session state.
decisions:write
Profiles: v2, commander, planner, executor, observer
Request:
orgx_recommend read-only
Recommend next work, summarize operator-chronicle/morning-brief signals, or read prioritization context. USE WHEN: user asks what to do next, wants a brief, asks what changed yesterday/week/30 days, or needs priority guidance. mode=morning_brief returns the operator chronicle when available. NEXT: present the recommendation, then use orgx_act, orgx_write, or orgx_spawn only after the user confirms an action. DO NOT USE WHEN: the user already specified the action; use orgx_act or orgx_write.
Recommendation mode; default next_action Options:
next_action, morning_brief.Reporting period for mode=morning_brief; default 30d Options:
day, week, 30d.Recommendation scope type Options:
workspace, initiative, workstream, milestone, task.Scoped entity ID
Workspace UUID
Maximum recommendations
Optional bootstrap/session identifier
decisions:read, agents:read, initiatives:read, memory:read
Profiles: v2, memory, commander, observer
Request:
orgx_search read-only
Find OrgX entities, decisions, artifacts, and memory. Also known as: Search OrgX, find initiative ID, list work, browse OrgX. USE WHEN: browsing work, searching memory, finding IDs, or listing related records. NEXT: use orgx_inspect for one selected result or orgx_recommend when the user asks what to do next. DO NOT USE WHEN: you already know the exact entity and need full context; use orgx_inspect.
Search query for memory or title/text matching
Optional entity type filter, such as task, milestone, decision, artifact, or initiative
Optional status filter
Optional initiative UUID scope
Optional workspace UUID scope
Maximum records to return
Optional compact field list
Optional bootstrap/session identifier
initiatives:read
Profiles: v2, memory, commander, planner, executor, observer
Request:
orgx_spawn
Guard, estimate, classify, spawn, or hand off specialist agent work. Per-action requirements: spawn from an existing task REQUIRES task_id; ad-hoc spawn REQUIRES title + instructions and should include agent_type. handoff REQUIRES task_id + agent_type. guard REQUIRES agent_type. classify REQUIRES title or task_id. action=“estimate” REQUIRES title or task_id, returns candidate routes/cost context, and runs without dispatching work. Routing policy: omit model_tier/provider/model for OrgX auto-routing. Set model_tier, provider, model, budget_mode, or max_cost_usd only when the user, policy, or verification plan constrains routing. For controlled reliability validation, use model_tier=“standard” and budget_mode=“cheapest_valid”. USE WHEN: explicitly delegating work to an OrgX agent or checking if delegation is allowed. NEXT: use orgx_inspect or orgx_search to monitor the delegated work, then orgx_submit_receipt for proof. DO NOT USE WHEN: only creating a task row; use orgx_write.Spawn operation. Defaults to “spawn”. Use estimate for pre-spawn cost/routing context without dispatching work. See top-level description for per-action required fields. Options:
guard, estimate, spawn, handoff, classify.Task title. REQUIRED for ad-hoc spawn (action=spawn without task_id) or action=classify without task_id. Used as the human-readable label of the spawned task.
Existing task UUID. REQUIRED for action=handoff. REQUIRED for action=spawn when spawning work for an already-created task. Either task_id or title (with instructions) must be provided for action=spawn.
Optional initiative UUID to scope the spawned task. Inferred from task_id when omitted.
Optional workspace UUID to scope the spawned task. Defaults to the MCP session’s workspace.
Target agent type/domain (e.g. “engineering”, “marketing”, “design”). REQUIRED for action=guard or action=handoff. Strongly recommended for action=spawn so the work routes to the right specialist.
Delegation instructions for the agent. REQUIRED for action=spawn when spawning ad-hoc (without task_id). Used to override the task description for action=handoff.
Optional model tier override. Omit to let OrgX auto-route from task complexity. Legacy local/sonnet/opus are accepted for older clients. Options:
standard, balanced, precision, local, sonnet, opus.Optional exact model identifier when the user explicitly selects one. Otherwise OrgX resolves the model from task, tier, provider, policy, and budget.
Optional provider preference. Use auto unless the user asks for a specific provider or a cost comparison selects one. Options:
auto, openai, anthropic, openrouter, groq, local.Optional budget posture override. Use cheapest_valid for controlled validation runs while reliability is being proven. Options:
cheapest_valid, balanced, highest_quality.Optional per-task hard cost ceiling in USD. If the estimate exceeds this, OrgX should block, downgrade, or request approval before dispatch.
Optional client-supplied idempotency key for safe retries. Same key returns the same spawn result without re-running.
Optional bootstrap/session identifier returned by orgx_bootstrap.
agents:write
Profiles: v2, commander, executor
Request:
orgx_submit_receipt
Submit a durable receipt (proof, outcome, quality, attribution, or learning) anchored to an OrgX entity or artifact. Required: receipt_type + summary. Strongly recommended: one anchor (entity_type+entity_id OR artifact_id), artifact_type, business_outcome, agent_type, AND at least one verifiable URL inside evidence. Recognized receipt_type: “proof” (completion proof), “outcome” (measurable result), “quality” (review/score), “attribution” (credit link to revenue/value), “learning” (distilled lesson). Custom keys also accepted. Recognized evidence shapes (mix and match): { prs: string[] } — GitHub PR URLs. { deploys: string[] } — deployment URLs. { test_runs: string[] } — CI run URLs. { metrics: { name, value, unit? }[] } — quantitative outcomes. { links: string[] }, { notes: string } — supporting URLs/text. Pass idempotency_key when retrying — server deduplicates. USE WHEN: closing the loop on agent work with provenance. For founder/team work, receipts should prove the practical artifact and its business outcome, not just say the agent finished. NEXT: orgx_recommend or orgx_search to find the next priority. DO NOT USE for telemetry — use orgx_emit_activity.Workspace UUID. Defaults to the MCP session’s workspace when omitted.
Related entity type (initiative, workstream, milestone, task, decision). Required if no artifact_id is provided — pair with entity_id.
Related entity UUID. Required when entity_type is provided.
Receipt category key. Recognized values: “proof”, “outcome”, “quality”, “attribution”, “learning”. Custom domain-specific keys are also accepted.
One-sentence human-readable description of what the receipt proves (e.g. “Merged PR #142 unblocking the auth refactor”).
Structured evidence payload. Recognized shapes: { prs: string[] }, { deploys: string[] }, { test_runs: string[] }, { metrics: { name, value, unit? }[] }, { links: string[] }, { notes: string }. See top-level description for full list. At least one verifiable URL is strongly recommended.
Related artifact UUID to anchor the receipt to. Alternative to entity_type+entity_id when the proof lives in OrgX as an artifact.
Artifact type this receipt proves. Preferred founder/team examples: orchestration.next_initiative, eng.pull_request, eng.deploy_proof, eng.structured_blocker, sales.strategy, sales.icp_offer_sequence, sales.send_plan, sales.conversion_gates, marketing.launch_asset, marketing.channel_hypothesis, marketing.positioning_brief, marketing.proof_distribution_plan, marketing.interview_pr_plan, product.customer_discovery, product.prd, product.pricing_hypothesis, product.decision_record, design.audit, design.component_spec, design.token_package, ops.operator_brief, ops.runbook, ops.budget_envelope, ops.incident_status, proof.link.
Agent/domain that produced the receipt, such as engineering, sales, product, design, operations, marketing, or orchestrator.
Business outcome advanced by the receipt.
Whether the artifact verification passed, failed, was blocked, or was not run. Options:
passed, failed, blocked, not_run.Optional OrgX loop validation rung this receipt proves. Required for promotion-grade loop validation receipts. Options:
schema_discovery, dry_run_spawn, single_domain_fixture, cross_domain_initiative_fixture, overnight_canary, calibrated_auto_route_expansion.Set true when this receipt should be evaluated against the OrgX loop reliability validation ladder.
Model tier used for the run being receipted. For validation rungs before calibrated expansion, use standard. Options:
standard, balanced, precision, local, sonnet, opus.Budget posture used for the run being receipted. For validation rungs before calibrated expansion, use cheapest_valid. Options:
cheapest_valid, balanced, highest_quality.Per-task or canary spend cap used during the validation run, when known.
Strongly recommended client-supplied idempotency key. Submitting the same key twice will not create a duplicate receipt.
Optional bootstrap/session identifier returned by orgx_bootstrap.
initiatives:write
Profiles: v2, memory, commander, executor
Request:
orgx_write
Create or update one OrgX record (snake_case fields). Operations: create (default) uses per-type fields; update REQUIRES id + fields. Create requirements: workspace name/title; initiative title/name + workspace_id + goal_ids when the workspace enforces primary objectives; workstream title + initiative_id; milestone title + workstream_id; task title + workstream_id + milestone_id when the workspace requires backlog milestones; decision title; artifact target + artifact_type + artifact_url/external_url; blocker run_id + metadata.description; skill/studio records title. Retry behavior: pass idempotency_key on creates. A key match returns the same UUID as an idempotent replay, without creating a duplicate. Initiative gotchas: priority only accepts low|medium|high|urgent, not portfolio labels such as active/critical/maintenance/hold; put portfolio urgency in metadata/priority_rank. due_date is not accepted on initiative create in current workspaces; put target dates in metadata until a typed initiative schedule field exists. USE WHEN: adding/editing records. NEXT: orgx_act to launch/complete the record. DO NOT USE for lifecycle changes — use orgx_act or orgx_attach.Write operation. Defaults to “create”. Set “update” (with id + fields) to patch an existing entity. Options:
create, update.Entity type to write: workspace, task, milestone, decision, artifact, skill, blocker, studio_brand, studio_content, initiative, workstream, or objective. See top-level description for per-type required fields.
REQUIRED when operation=“update”. Target entity UUID to patch.
REQUIRED on create (provide either “title” or “name” — they are aliases). Display title of the new entity.
Alternative to “title” on create. REQUIRED on create when “title” is not provided.
Short description shown in lists and previews. Recommended on create.
Longer-form description used in detail views.
REQUIRED when operation=“update”. Map of entity fields to patch (only include fields you want to change).
Parent initiative UUID. REQUIRED when type=“workstream”. Optional context for tasks/milestones/artifacts to associate them with an initiative.
Parent workstream UUID. REQUIRED when type=“milestone” or type=“task”.
Parent milestone UUID for tasks. Some workspaces require an explicit backlog milestone under the workstream; create/resolve that milestone first instead of relying on auto-resolution.
Workspace UUID. REQUIRED when the MCP session does not already carry workspace context (resolve via list_entities type=command_center or orgx_inspect type=workspace).
Objective UUIDs for initiative/workstream/milestone/task creation. REQUIRED when the workspace enforces a primary objective. Resolve via orgx_inspect type=objective.
Record urgency. Only low|medium|high|urgent are accepted here; do not send portfolio/live labels such as active, critical, maintenance, or hold. “urgent” is normalized to “high” server-side. Options:
low, medium, high, urgent.Due date as YYYY-MM-DD for supported entity types. Do not send due_date when type=“initiative”; put initiative target dates in metadata.
Initial workflow status; common agent aliases such as “active” are normalized per entity type (“active” → “in_progress”).
REQUIRED when type=“artifact”. Entity type to attach the artifact to (initiative, workstream, milestone, task, or decision).
REQUIRED when type=“artifact” (unless task_id is provided). UUID of the entity to attach the artifact to.
Shortcut for attaching an artifact directly to a task. Use instead of entity_type+entity_id when type=“artifact” and the target is a task.
REQUIRED when type=“artifact”. Artifact type code. Preferred founder/team examples: orchestration.next_initiative, eng.pull_request, eng.deploy_proof, eng.structured_blocker, sales.strategy, sales.icp_offer_sequence, sales.send_plan, sales.conversion_gates, marketing.launch_asset, marketing.channel_hypothesis, marketing.positioning_brief, marketing.proof_distribution_plan, marketing.interview_pr_plan, product.customer_discovery, product.prd, product.pricing_hypothesis, product.decision_record, design.audit, design.component_spec, design.token_package, ops.operator_brief, ops.runbook, ops.budget_envelope, ops.incident_status, proof.link. Custom codes remain accepted.
Internal artifact URL (e.g. /api/artifacts/…). Either artifact_url or external_url is required when type=“artifact”; preview_markdown alone is not accepted.
External artifact URL (https://). Either artifact_url or external_url is required when type=“artifact”; preview_markdown alone is not accepted.
Optional inline markdown preview shown with the linked artifact. Supporting context only; it does not replace artifact_url/external_url.
REQUIRED when type=“blocker”. Agent run UUID the blocker applies to.
Optional agent run step UUID for blocker creation.
Blocker category/type when type=“blocker” (e.g. “missing_input”, “permission”, “external_dependency”).
Blocker resolution text when known. Used to mark a blocker as resolved.
Initiative live-link visibility. Only applies when type=“initiative”. Options:
private, public.Shortcut to publish an initiative live link (sets live_visibility=“public”). Only applies when type=“initiative”.
When true, public live-link visitors see the initiative title. Only applies when type=“initiative” with live_visibility=“public”.
Free-form object for type-specific metadata. Schema varies per entity type (e.g. for skills: { capabilities, guardrails, channels }; for studio_brand: { tokens, voice, exemplars }).
Workspace tagline when type=“workspace”.
Workspace identity narrative when type=“workspace”.
Workspace identity metrics when type=“workspace”.
Workspace roadmap URL when type=“workspace”.
Workspace source links when type=“workspace”.
For workspace create, make the new workspace active. Defaults true.
Strongly recommended client-generated idempotency key for safe retries. Same key returns the same result without creating a duplicate.
Optional bootstrap/session identifier returned by orgx_bootstrap.
initiatives:write
Profiles: v2, commander, planner, executor
Request:
Other
10 tools for other operations.check_execution_readiness read-only
Check whether the workspace has the credentials needed to dispatch agent work, BEFORE spawning. USE WHEN: about to call orgx_spawn / spawn_agent_task and you want to confirm execution credentials exist — avoids dispatching a run that fails on missing keys. NEXT: if not ready, surface what is missing and resolve it; if ready, proceed to orgx_spawn. Read-only; dispatches nothing.
Workspace UUID to check. Defaults to the MCP session workspace.
initiatives:read
Profiles: all
Request:
configure_outcome_type
Create or approve a workspace outcome type before recording custom baseline, audit, or quality-gate outcomes. Use this when record_outcome returns an unknown outcome type recovery hint.Workspace ID
CamelCase alias for workspace_id
Outcome type key, normalized server-side to snake_case
Human-facing outcome type label
CamelCase alias for display_name
Measurement unit for this outcome type Options:
usd, hours, count, percent. Default: "count".How the value should be interpreted by ROI/proof loops Options:
revenue, time_saved, risk_reduced, quality_improved. Default: "quality_improved".CamelCase alias for value_semantics Options:
revenue, time_saved, risk_reduced, quality_improved.initiatives:write
Profiles: all
Request:
get_operator_chronicle read-only
Read the operator chronicle: decision chronology, yesterday/week/30-day rollups, reportingNarrative.briefMarkdown, artifacts, PR receipts, active initiatives, goals, top priorities, velocity, and reporting gaps for a workspace. USE WHEN: user asks what changed yesterday, this week, or this month; asks for decision chronology, top priorities, PR velocity, artifacts, goals, or what OrgX is missing. NEXT: present reportingNarrative.briefMarkdown first, then use reportingNarrative.nextAction and topPriorities for drill-down. Read-only.
Workspace UUID to scope the chronicle.
Deprecated alias for workspace_id.
Reporting window. day=24h, week=7d, 30d=30 days. Options:
day, week, 30d.initiatives:read
Profiles: v2, commander, observer
Request:
manage_lifecycle
Pause, resume, retry, or cancel any node in the work hierarchy — an initiative, workstream, milestone, task, or run. Propagates to descendant tasks + active runs. USE WHEN: the user wants to halt, restart, or re-run part of an initiative. resume/retry re-dispatch the work; pause/cancel stop active runs. NEXT: get_initiative_pulse to confirm. DO NOT USE: to mark a task done — use entity_action instead.Which hierarchy node to act on Options:
initiative, workstream, milestone, task, run.UUID of the node
Lifecycle action to apply Options:
pause, resume, retry, cancel.agents:write
Profiles: all
Request:
orgx_emit_execution_graph
Emit the deterministic execution graph + trust ledger for the active run: nodes (claimed status + verification evidence), depends_on edges, and trust events. OrgX derives false-completion, hallucinated-receipt, and dependency-violation signals from this and surfaces them on /live. USE WHEN: reporting the structured shape of multi-step work — NOT a single progress line (use orgx_emit_activity for that). NEXT: continue work; re-emit as the graph advances (idempotent per run + graph fingerprint). DO NOT USE to mark entities complete — telemetry only.Initiative UUID
Existing run UUID
Required when run_id is not provided
Required when run_id is not provided Options:
openclaw, codex, claude-code, chatgpt, cursor, web-ui, api.Runtime provenance used by /live to bucket chokepoints
Optional human-readable rollup of the graph state
Execution graph nodes; each carries claimed status + optional verification
depends_on edges between node ids
Explicit trust-ledger events such as authority_exceeded that OrgX cannot infer from graph shape
Optional structured metadata to attach to the execution-graph emission
orgx_free_audit read-only
Run a free autonomy benchmark from trust, proof, ROI, and workspace signals. Returns Proof Score, Context Debt, Autonomy Maturity, ROI Visibility, and next recommendations without starting an autonomous session or consuming agent credits.
Workspace ID to audit
Agent type to benchmark trust against Default:
"orchestrator".ROI attribution period used for ROI Visibility Options:
7d, 30d, 90d. Default: "30d".Include raw upstream signal payloads for debugging and verification Default:
false.agents:read, initiatives:read
Profiles: all
Request:
orgx_poll_question read-only
Read the durable answer receipt for a question previously forwarded with orgx_request_question. USE WHEN: an external client is paused for human input. NEXT: when resolved=true and continuation.should_resume=true, resume the preserved source session with answer and resolution_context; when false, remain paused without creating another question. Read-only.
Question/decision UUID returned by orgx_request_question
orgx_request_question
Pause at a safe checkpoint and forward a contextual question to the initiative owner. OrgX renders the right single-select, multi-select, free-text, or confirmation control and preserves the source run/session so work can resume with the exact answer. USE WHEN: Claude Code AskUserQuestion, Codex request_user_input, Cursor, OpenClaw, or another client needs human judgment to continue. NEXT: poll orgx_poll_question with the returned question_id; continue only when continuation.should_resume is true. DO NOT USE: for runtime failures with an automatic retry path — report a chokepoint instead.Initiative UUID
Stable key for safe retries of the same question
The concrete question the owner must answer
What led here, what is preserved, and what the answer changes
Pre-allocated answers for single-select or multi-select questions
Answer control; inferred from options when omitted Options:
single_select, multi_select, free_text, confirmation.Whether this answer gates the linked work; default true
Owner attention urgency; default medium Options:
low, medium, high, urgent.Affected workstream UUID
Existing run UUID
Required with source_client when run_id is absent
Originating client; required when run_id is absent Options:
openclaw, codex, claude-code, chatgpt, cursor, web-ui, api.Runtime/provider provenance for the paused execution
Claude, Codex, Cursor, or other client session identifier
Originating tool, e.g. AskUserQuestion or request_user_input
Originating question/tool-call event identifier
Additional non-secret client continuation references
Additional non-secret structured context
queue_action
Pin, unpin, or skip a workstream in the Next Up queue. USE WHEN: user wants to force-prioritize or deprioritize a workstream. action=pin to force-top, action=unpin to remove pin, action=skip to temporarily deprioritize. Requires initiatives:write.Queue operation Options:
pin, unpin, skip.Initiative UUID
Workstream UUID
Workspace UUID
Deprecated alias for workspace_id
Position among pinned items, 0=top (pin only)
Skip duration in minutes, default 60, max 10080 (skip only)
initiatives:write
Profiles: all
Request:
scoring_config
Read or update scoring engine configuration. USE WHEN: user asks about scoring setup, wants to toggle scoring, adjust weights, or change active signals. action=get to read, action=update to modify. NEXT: Run recommend_next_action to see effects in the preferred workflow, or score_next_up_queue if you need raw queue scoring. Read-only for get, requires initiatives:write for update.get=read config, update=modify settings Options:
get, update.Workspace UUID
Deprecated alias for workspace_id
Enable/disable scoring_v2 (action=update only)
Signal names to activate (action=update only)
Partial weight overrides e.g. { priority: 300 } (action=update only)
decisions:read, agents:read, initiatives:read, memory:read
Profiles: all
Request:
Deprecated Tools
Next Steps
MCP Protocol
Protocol implementation details.
Agent Ops
Get started with agent workflows.