ck_review_submit
Submit a plan, diff, or completion packet for human review and execution gating. Creates a review record and returns a review_id and browser URL for approval or denial.
Instructions
Submit a governed plan, diff, or completion packet for human review and execution gating. Write operation — creates a review record and returns a review_id and browser URL. review_type controls what is being submitted: plan (before implementation), diff (before merging), or completion (task done). submission_body is the full content: plan text, diff, or completion description. For iterative plan refinement, pass previous_review_id and plan_phase (ticket → research_packet → design_options → narrowed_decision → implementation_plan → code_backed_plan). The plan-quality scorer evaluates structured fields, not just submission_body — populate research_summary, options_considered, selected_option, rejected_options, implementation_steps, validation_plan, code_snippets, alignment_context, consulted_roles, codebase_findings, prior_art_summary, agent_spec_id, task_spec_id, agent_role, task_scope, out_of_scope, business_rules, domain_terms, allowed_actions, prohibited_actions, robustness_requirements, linked_policy_packs, linked_benchmark_suites, promotion_gates, allowed_semantic_changes, forbidden_semantic_changes, invariant_boundaries, requires_reapproval_if, harness_quality_checks, and scope_estimate for a strong score. Returns review_id, status (pending), and a URL where the human reviewer can approve or deny. After submission, poll ck_review_status until the decision is approved or denied before proceeding. Use ck_review_feedback (human-facing) to record a decision on an existing review.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| title | No | Human-readable title for display and search. | |
| task_id | No | Task identifier within the session for scoped operations. | |
| metadata | No | Arbitrary key-value metadata for extensibility and audit context. | |
| agent_role | No | Reviewed role label such as support agent, code reviewer, deployment agent, or sales assistant. | |
| plan_phase | No | Current phase of plan refinement. | |
| session_id | No | Unique session identifier for correlating findings, proofs, budget, and audit trail. | |
| task_scope | No | What the agent or task is expected to accomplish under this plan. | |
| annotations | No | Structured key-value annotations for machine-readable metadata. | |
| review_type | No | Type of review being submitted or queried (plan, diff, or completion). | |
| domain_terms | No | ||
| out_of_scope | No | ||
| project_root | No | Absolute path to the project root this submission belongs to. Pass it when one MCP server serves several repos (e.g. an IDE-wide CONTROLKEEL_PROJECT_ROOT) so the review pins to the intended project instead of inheriting the server's working directory. | |
| submitted_by | No | Identity of the submitter for audit trail. | |
| task_spec_id | No | Stable identifier for the task-level behavior contract this plan is implementing. | |
| agent_spec_id | No | Stable identifier for the agent role or task contract this plan is implementing. | |
| code_snippets | No | ||
| business_rules | No | ||
| feedback_notes | No | Freeform feedback notes from the reviewer. | |
| scope_estimate | No | ||
| allowed_actions | No | ||
| consulted_roles | No | ||
| promotion_gates | No | ||
| selected_option | No | The chosen approach with rationale. | |
| submission_body | Yes | Full submission content: plan text, diff, or completion description. | |
| validation_plan | No | ||
| rejected_options | No | ||
| research_summary | No | Summary of research performed before this submission. | |
| alignment_context | No | ||
| codebase_findings | No | ||
| prior_art_summary | No | Summary of prior attempts or related work. | |
| options_considered | No | ||
| previous_review_id | No | Reference to a prior review for iterative refinement. | |
| prohibited_actions | No | ||
| linked_policy_packs | No | ||
| implementation_steps | No | ||
| invariant_boundaries | No | System invariants and boundaries that must remain true during execution. | |
| harness_quality_checks | No | Agent-harness quality checks such as context hygiene, proof completeness, rollback safety, and compaction fidelity. | |
| requires_reapproval_if | No | Conditions that require human re-approval before continuing. | |
| linked_benchmark_suites | No | ||
| robustness_requirements | No | ||
| allowed_semantic_changes | No | Semantic behavior changes explicitly approved for this plan. | |
| persona_or_actor_context | No | User role, customer tier, permission state, or operating context relevant to behavior. | |
| forbidden_semantic_changes | No | Semantic behavior changes the agent must not introduce without a new review. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| title | No | ||
| status | No | ||
| review_id | No | ||
| review_url | No |