.claude/skills/codex-lifecycle-orchestration/SKILL.md
End-to-end lifecycle orchestration with a single Codex agent. Replaces subagent workflow coordination while preserving scale-based phases, stop points, and quality gates.
npx skillsauth add ssaattww/excelreport codex-lifecycle-orchestrationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill follows the non-entry execution contract standard.
Required contract_extensions: lifecycle_scale, phase.
See codex-execution-contract.md and non-entry-execution-contract-template.md for full rules.
input:
objective: "Run medium lifecycle orchestration"
contract_extensions: { lifecycle_scale: "medium", phase: "design" }
output:
status: "completed"
quality_gate:
gate_id: "lifecycle-phase-check"
gate_type: "implementation"
trigger: "post-phase validation"
criteria:
- "Phase-specific checks are complete"
- "Lifecycle contract fields remain aligned"
result: "pass"
evidence:
- "Phase checks passed"
blockers: []
branching:
on_pass: "handoff"
on_fail: "revise"
max_cycles: 2
contract_extensions: { lifecycle_scale: "medium", phase: "design" }
Emit structured output compliant with codex-execution-contract.md.
Always include required fields: status, summary, changed_files, tests, quality_gate, blockers, next_actions.
Include required extension echo fields in contract_extensions: lifecycle_scale, phase.
Treat missing required fields or invalid status values as contract violations and regenerate before handoff.
Template reference: non-entry-execution-contract-template.md.
Validate required input fields from non-entry-execution-contract-template.md (objective, scope, constraints, acceptance_criteria, allowed_commands, sandbox_mode) before proceeding.
See references/legacy-subagent-mapping.md for one-to-one replacement of subagent responsibilities.
| Scale | File count | PRD | ADR | Design Doc | Work Plan | |---|---|---|---|---|---| | Small | 1-2 | Update if exists | Optional | Optional | Simplified | | Medium | 3-5 | Update if exists | Conditional | Required | Required | | Large | 6+ | Required | Conditional | Required | Required |
ADR is conditional when architecture, technology choice, or data flow changes.
[Stop: pre-design-approval] + [Approve: design-approval].[Stop: pre-design-approval] + [Approve: design-approval].[Stop: pre-design-approval] + [Approve: design-approval] for design handoff.[Stop: pre-implementation-approval] + [Approve: implementation-start].[Stop: pre-design-approval] + [Approve: design-approval].[Stop: pre-implementation-approval] + [Approve: implementation-start].[Stop: pre-implementation-approval] + [Approve: implementation-start].task-analyzer, implementation-approachdocumentation-criteriacoding-principles, testing-principlesai-development-guideintegration-e2e-testingUse canonical markers [Stop: <Gate Name>] and classify each stop as approval_gate or escalation_gate.
Use normalized stop payloads with status, full gate record fields required by stop-approval-section-template.md (including gate_name, gate_type, trigger, ask_method, required_user_action, resume_if, fallback_if_rejected), and quality_gate.result.
approval_gate resumes only with explicit user approved: true; escalation_gate resumes only after user direction/reroute.
Respect the batch boundary: emit [Stop: pre-implementation-approval] before any autonomous implementation loop.
Set max_revision_cycles: 2; if exceeded, escalate and require human intervention.
Template reference: ../workflow-entry/references/stop-approval-section-template.md.
Stop points for this skill:
[Stop: pre-design-approval] (approval_gate)[Stop: pre-implementation-approval] (approval_gate)[Stop: high-risk-change] (approval_gate)[Stop: requirement-change-detected] (escalation_gate)[Stop: quality-gate-failed] (escalation_gate)[Stop: revision-limit-reached] (escalation_gate)Orchestrator ownership: this skill emits quality_gate evidence and decides pass/fail/blocked branching.
Use canonical gate fields (gate_id, gate_type, trigger, criteria, result, evidence, blockers, branching) per template.
Normalize local outcomes to result: pass|fail|blocked before handoff.
If result: blocked, emit [Stop: quality-gate-failed] and pause autonomous flow.
Use gate-type branching.max_cycles limits from quality-gate-evidence-template.md (default 2, stricter limits win).
Map gate_type based on the current lifecycle phase:
gate_type: documentgate_type: consistencygate_type: implementationdatabases
Unified deterministic entry for workflow requests. Centralizes routing with stop/approval and sandbox controls.
development
Sends commands to another tmux pane. Use when requests include phrases like "run it in another pane," "send via tmux," or "ask Claude Code to execute."
tools
Use when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automated editing
development
Single-agent execution loop for implementation tasks. Replaces task-executor/quality-fixer subagent cycle with direct Codex skill-driven implementation and quality gates.