orgx_attach_artifact
ChatGPTRegister a named deliverable and provenance on an OrgX work target. A durable artifact URL is required. May trigger automatic evaluation using external model providers and incur model costs. Registered proof awaits review; attaching does not approve or complete work.
orgx_cancel_work
ChatGPTCancel a hierarchy node through OrgX lifecycle control, propagating to descendant work and stopping active runs. Does not mark work complete.
orgx_capture_decision
ChatGPTCapture decision text, rationale, and context for human review. Returns its authoritative review state. Capturing a decision does not approve it or establish an accepted historical judgment.
orgx_check_execution_readiness
ChatGPTRead whether the authorized workspace has execution credentials and configuration before dispatching agent work. Never dispatches a run.
orgx_complete_plan
ChatGPTSave final markdown and complete the planning session only if its expected revision is current. Optionally attach the completed plan to typed work targets; attachment failures are reported separately after completion. Does not start execution. After an uncertain response, read the plan before repeating the completion.
orgx_complete_work_with_proof
ChatGPTRegister typed proof, verify completion requirements, and apply the allowed work completion transition. May return proof recorded with completion blocked or awaiting review. Verification and human acceptance remain separate facts.
orgx_create_initiative
ChatGPTCreate one initiative in its initial state without launching work or publishing a live link. Supply objectives when workspace policy requires them.
orgx_create_initiative_hierarchy
ChatGPTAtomically commit a deterministic initiative hierarchy in OrgX using a stable retry key. Creates work records without launching agents. Visibility defaults to private; source evidence is retained.
orgx_create_milestone
ChatGPTCreate a milestone under a workstream. Does not start or complete work.
orgx_create_task
ChatGPTCreate one task under a workstream. Supply a milestone when workspace backlog policy requires one. Does not dispatch the task.
orgx_create_workstream
ChatGPTCreate a workstream under an existing initiative. Does not dispatch agents or start the workstream.
orgx_estimate_agent_task
ChatGPTEstimate routing and cost for an existing task or a proposed task. Never creates a task, consumes a dispatch allowance, or starts a run.
orgx_get_agent_status
ChatGPTRead active and optionally idle agents in the workspace or initiative. Successful calls record metered MCP allowance usage without changing work records.
orgx_get_initiative_progress
ChatGPTRead hierarchy progress, health, blockers, and review gates for one initiative. Successful calls record metered MCP allowance usage without changing work records.
orgx_get_next_actions
ChatGPTRead actionable priorities and blockers for the selected work scope. The compatibility implementation records metered MCP allowance usage without changing work records.
orgx_get_operation_status
ChatGPTRead a durable work command, run, or decision. Prefer operation_id returned by a prior operation; compatibility callers can supply kind and id together. Never starts or settles work.
orgx_get_operator_brief
ChatGPTRead workspace reporting, decision chronology, progress, proof, priorities, and gaps for a day, week, or 30-day window.
orgx_get_receipt_review_queue
ChatGPTRead the workspace calls waiting for a person: outcome, work type, entity mapping, area, links and proposed criteria. Results are suggestions with confidence, not verified facts. Human outcome judgments and criterion confirmations are made in the OrgX review UI; this tool cannot resolve them.
orgx_get_work_receipt
ChatGPTRead a receipt by its ledger UUID or opaque producer receipt ID in one workspace. Returns bounded evidence and criteria, producer claims, the separately recorded human outcome, related workstream and uncertainty. Reading cannot verify or accept work. The full document remains in OrgX.
orgx_get_workspace_context
ChatGPTRead the authorized workspace, optional initiative context, and typed references. Does not select a workspace or create a session.
orgx_handoff_task
ChatGPTReassign an existing task to a specialist through the canonical handoff and dispatch lifecycle. Can incur model costs and connected-service effects; requires agent and initiative write grants.
orgx_inspect
ChatGPTRead the current state and linked context of one OrgX work record, decision, artifact, or plan session before acting. Returns its owners, decisions, acceptance checks, and proof where available. Related records are available through orgx_search and priorities through orgx_get_next_actions. Does not mutate the record.
orgx_launch_initiative
ChatGPTLaunch an existing initiative after OrgX approval, budget, and readiness checks. May dispatch agents, incur model costs, and interact with connected services.
orgx_list_pending_decisions
ChatGPTRead pending decisions awaiting human review in the selected workspace or initiative. Does not resolve decisions or dispatch work.
orgx_list_work_receipts
ChatGPTRead one bounded workspace ledger view: receipts (default), workstreams, or review. Receipt search accepts free text and exact filters: outcome, verification, accepted, type, area, repo, actor, ws, entity, initiative, pr, file, since, until, conf, unmet and status, within the 120-day window. Workstreams retain receipt rollups and confirmed mappings; review returns pending human calls and criterion proposals. Results distinguish producer claims from human judgments. Follow validated next_calls; no view resolves judgments, runs model inference, or changes work.
orgx_open_artifact_review
ChatGPTRead an artifact review packet and open the review widget. Supply an artifact ID, scope the next pending artifact to an entity, or use the authenticated workspace pending queue.
orgx_open_decision_review
ChatGPTOpen the complete human review surface for one pending decision. The person selects the ruling; this operation does not approve, reject, or resume work.
orgx_panel_snapshot
ChatGPTApp-only: the OrgX panel in the ChatGPT sidebar and beside a conversation. Returns what needs your decision (most urgent first), the review packet for one of them, the last output a person accepted, and, when asked, the work agents are running. USE WHEN: the OrgX panel opens or refreshes. NEXT: the panel shows Approve and Send back when the decision can be settled there, otherwise it links to the decision in OrgX. DO NOT USE: from a model; models read decisions with orgx_search. Read-only. Work receipts are read with orgx_list_work_receipts and orgx_get_work_receipt.
orgx_pause_work
ChatGPTPause a hierarchy node through OrgX lifecycle control, propagating to descendant work and stopping active runs. Does not mark work complete.
orgx_read_plan
ChatGPTRead a planning draft by session UUID or plan URI. If omitted, read the most recent active plan in the authenticated workspace.
orgx_record_plan_edit
ChatGPTAppend a change or feedback summary to the durable plan history. This records guidance only; it does not replace the markdown content or complete the plan.
orgx_request_independent_artifact_review
ChatGPTQueue an independent evaluator for a completed artifact. OrgX chooses the evaluator and persists the assessment. May incur model costs or use external providers; the producing agent cannot choose the score.
orgx_resume_work
ChatGPTResume paused work through canonical OrgX recovery, preserving attempt lineage and redispatching where permitted. Can incur costs and connected-service effects.
orgx_retry_work
ChatGPTRetry eligible work through canonical OrgX recovery, preserving attempt lineage and redispatching where permitted. Can incur costs and connected-service effects.
orgx_save_plan
ChatGPTSave the full markdown draft in OrgX only if the required expected revision is current. Optionally record an edit summary. Does not complete planning or dispatch work.
orgx_search
ChatGPTSearch authorized OrgX records. To list records, set type and omit query. Query is text, not a command: use status and created_from/created_to for filters. Typed search pages collections; mixed search covers a bounded memory relevance window. Request logs, source files, and external systems are not searched here. Mixed relevance searches record metered MCP allowance usage without changing work records. Follow next_call for pagination. Full context is available through orgx_inspect, priorities through orgx_get_next_actions, and receipt ledger searches through orgx_list_work_receipts.
orgx_start_agent_task
ChatGPTDispatch one existing OrgX task under budget, deadline, acceptance, eligibility, and permission policy. Can incur model costs and act on connected services within the declared effect constraints.
orgx_start_plan
ChatGPTCreate a durable planning draft with a title and optional initial markdown. An optional retry key replays the same unedited draft; conflicting content or a draft edited since creation returns a conflict. Does not create executable work or dispatch agents.
orgx_submit_work_receipt
ChatGPTImport one portable Agent Work Receipt v0.1 or v0.2 into the authenticated workspace ledger. Preserves the full document, including identified criteria, evidence, provenance, lineage and review extensions. Producer outcome, verification and acceptance are claims; import does not verify, accept or complete work. The canonical import enforces document limits, reserved harness names, workspace quota and content-bound idempotency. Never falls back to a legacy write after rejection.
orgx_update_work
ChatGPTApply a nonempty typed content patch to initiative, workstream, milestone, or task. Permits text, priority, dates, objectives, notes, and tags; cannot change workflow status, approvals, ownership authority, or execution policy.
orgx_validate_initiative_plan
ChatGPTValidate a deterministic hierarchy and source evidence in OrgX; return its digest and structural findings. Does not create records, launch agents, or invoke paid generation.
orgx_validate_work_receipt
ChatGPTCheck one portable Agent Work Receipt v0.1 or v0.2 for schema and reference conformance without storing it. Checks identified criteria, evidence references and known review extensions. Conformance validates the document, not whether its evidence is true or whether work was accepted. Import separately enforces hosted compatibility and admission policy.
orgx_widget_approve_artifact
ChatGPTWidget-only: record a person’s approval of the displayed artifact using its hidden, version-bound capability. Refuses stale, foreign, missing, or replayed capabilities. Applies OrgX review and continuation rules.
orgx_widget_decide
ChatGPTWidget-only: settles one pending item (a decision of any type, an agent-run approval, or an action awaiting approval) after the person clicks in the decisions widget or the OrgX panel. Sends what the item's widget_actions ask for: an option, several options, a typed answer or a reason. USE WHEN: the widget sends a click with its approval token from result metadata, which models never see. NEXT: the widget shows the item as settled. DO NOT USE: from a model or without a widget approval token; use approve_decision or reject_decision to point the person to where they decide.
orgx_widget_receipt_call
ChatGPTApp-only: records a person's call on a Work Ledger receipt from the OrgX panel: done, partly done, not done or blocked. The ledger keeps the latest call per receipt; it overrides the agent's own guess and teaches OrgX what done means. USE WHEN: the person presses a call on a receipt in the OrgX panel. DO NOT USE: from a model; models read receipts with orgx_get_work_receipt or orgx_list_work_receipts.
orgx_widget_request_artifact_changes
ChatGPTWidget-only: record a person’s requested changes to the displayed artifact using its hidden, version-bound capability and required feedback. Applies the canonical OrgX rework flow and may dispatch connected agent work.
orgx_widget_select_workspace
ChatGPTWidget-only: select the person’s authorized workspace for this connection. Updates private session continuity; does not create or change work.
resume_agent_run
ChatGPTWidget-only: resumes a paused or auto-closed agent run after the person clicks Resume in the agent-status widget. Flips status back to running, clears TTL auto-close markers, and appends a resume_history entry. USE WHEN: the agent-status widget sends a Resume click. NEXT: the widget polls orgx_command_status until the run settles. DO NOT USE: from a model (use manage_lifecycle level=run action=resume), or to restart a completed/failed/cancelled run — those are terminal.
