codex-conductorMethodical end-to-end software delivery orchestrator for Codex CLI with dual project modes (greenfield for new builds, brownfield for existing systems) and dual execution modes (autonomous and gated). Use when users want full lifecycle delivery with strict stage gates, progress tracking, per-step manual/automated testing, continuous docs updates, change-impact management, and a reusable AGENTS.md workflow for any coding agent.
Install via ClawdBot CLI:
clawdbot install ShalomObongo/codex-conductorCoordinate Codex as a disciplined delivery system, not a one-shot generator.
Select both:
project_modegreenfield: build from scratchbrownfield: onboard and modernize an existing systemexecution_modeautonomous: proceed automatically when gates passgated: pause at every gate for user approvalNo code without a spec. This is non-negotiable.
Before any implementation, a written spec must exist with:
The coding agent MUST NOT:
If spec is unclear → STOP and ask. Never guess.
See references/spec-driven-development.md for full spec templates and enforcement rules.
Never skip gates silently. Never implement without a spec.
Read these references before running:
references/spec-driven-development.md (MANDATORY FIRST - governs all work)references/planning-questionnaire.mdreferences/modes.mdreferences/gate-checklists.mdreferences/testing-matrix.mdreferences/manual-test-templates.mdreferences/codex-runbook.mdreferences/gate-prompts.mdscripts/agent_exec.pyreferences/research-playbook.md (if research_mode=true)Initialize project artifacts:
python scripts/init_project_docs.py --root <project-path> --mode <greenfield|brownfield>This creates/updates:
AGENTS.md (project workflow contract)docs/*.md planning/architecture/test/progress/change docs.orchestrator/status.json (machine-readable state).orchestrator/context.json (project/execution/research mode context)Before anything else, ask the user which coding agent to use (codex | claude | opencode | pi) and fallback agent.
Then ask all required questions from references/planning-questionnaire.md.
Minimum required answers:
project_modeexecution_modeIf research_mode=true, produce docs/research-notes.md and architecture recommendation before G2.
Must complete before G2:
Must complete before G2 (and authored by coding agent, not orchestrator):
Use gates G0 through G7 defined in references/gate-checklists.md.
Update gate state via script:
python scripts/gate_status.py set --root <project-path> --gate G3 --state PASS --note "slice-1 verified"Validate status schema:
python scripts/gate_status.py validate --root <project-path>Allowed states: PENDING | IN_PROGRESS | PASS | FAIL | BLOCKED.
By default, gate preconditions are enforced (sequence + mode-aware docs checks).
Use references/testing-matrix.md.
Mandatory checks per progression:
Also execute manual test scripts from references/manual-test-templates.md.
For each meaningful step:
docs/tasks.mddocs/progress.mddocs/change-log.mddocs/traceability.mddocs/test-results.mdFor user-requested changes, run:
python scripts/change_impact.py --root <project-path> --request "<change request>"Then complete all TODOs it emits in impacted docs.
Use PTY/background for long runs. Follow command patterns in references/codex-runbook.md.
Critical rule: each run executes ONE task, not a whole project in one prompt.
For G4, maintain docs/g4-task-plan.md checklist and process tasks one by one.
Generate gate-specific prompts with:
python scripts/generate_gate_prompt.py --gate <G1..G7> --agent <codex|claude|opencode|pi> --project-mode <greenfield|brownfield> --execution-mode <autonomous|gated> --research-mode <true|false> --task "<single task summary>" --spec-ref "<spec ref when applicable>"update_docs_step.py is now a fallback utility for recovery/manual bookkeeping only.
Primary expectation: the coding agent updates docs directly during each task.
Required loop:
Enforcement:
run_gate.py requires --spec-ref for G3/G4 tasks (implementation gates).run_gate.py requires coding agent + fallback agent context.--validate-cmd and/or --ui-review-note).--requires-browser-check must include --ui-review-note.status=PASS requires at least one --validate-cmd.status=PASS is blocked when --agent-dry-run is used.docs/g4-task-plan.md has no unchecked tasks.docs/validation-log.md.docs/agent-handoff.md.references/spec-driven-development.md.--auto-block-on-retry-exhaust auto-classifies gate as BLOCKED when retries are exhausted.Generate a quick status board:
python scripts/progress_dashboard.py --root <project-path>This summarizes current gate, completion %, blockers, and recent activity.
Run a single-task gate step with one command:
python scripts/run_gate.py --root <project-path> --gate G2 --agent codex --fallback-agent claude --project-mode brownfield --execution-mode gated --research-mode true --task "architecture baseline refined for API routing" --status IN_PROGRESS --validate-cmd "npm run -s typecheck" --ui-review-note "N/A for architecture-only task"Mark PASS only after all gate-level checklist items are complete:
python scripts/run_gate.py --root <project-path> --gate G2 --agent codex --task "architecture gate complete" --status PASS --validate-cmd "npm run -s typecheck"For web/UI tasks, require browser verification by OpenClaw agent:
python scripts/run_gate.py ... --requires-browser-check --ui-review-note "Verified login + CRUD manually in browser via OpenClaw browser tools"Package distributable skill artifact:
python scripts/package_skill.py --skill-dir . --out distAt completion provide:
docs/progress.md at 100%.orchestrator/status.jsonIf blockers remain, mark as PARTIAL_COMPLETE with explicit blockers and owners.
Generated Mar 1, 2026
A financial services company needs to migrate a monolithic legacy banking application to a microservices architecture with zero downtime. The orchestrator uses brownfield mode to analyze the existing system, create characterization tests, and manage phased migration with strict gates for each service extraction, ensuring compliance and security at every step.
A startup aims to build a new customer relationship management (CRM) platform from scratch. Using greenfield mode, the orchestrator guides spec-driven development from requirements gathering to deployment, with autonomous execution for rapid prototyping and gated approvals for critical features like data encryption and API design.
A healthcare provider requires updates to an existing patient records system to meet new regulatory standards. In brownfield mode, the orchestrator assesses the current architecture, identifies compliance gaps, and manages change-impact documentation, with gated execution for manual approvals at each security and privacy gate to ensure HIPAA compliance.
An online retailer wants to add a real-time inventory tracking feature to their existing platform. The orchestrator uses brownfield mode to onboard the system, create specs for the new feature, and execute tasks in gated mode, pausing for user approval at integration and testing gates to prevent disruptions to live operations.
An educational institution needs a new learning management system (LMS) built with custom analytics. In greenfield mode, the orchestrator facilitates spec creation, architecture design, and autonomous execution for non-critical components, while using gated mode for user approval on core features like data privacy and accessibility compliance.
Offer tailored software delivery consulting using the orchestrator to manage client projects, ensuring disciplined development with spec-driven approaches and gate-based approvals. Revenue is generated through hourly or project-based fees, with upsells for ongoing maintenance and training on the orchestrator workflow.
Provide the orchestrator as a managed service integrated into existing development platforms, charging subscription fees based on usage tiers. This model targets teams needing automated lifecycle management, with revenue from monthly subscriptions and premium support for advanced features like research mode and custom gate configurations.
Develop and sell training programs and certifications for teams to learn spec-driven development and gate-based project management using the orchestrator. Revenue comes from course fees, certification exams, and corporate training packages, helping organizations standardize their software delivery processes.
💬 Integration Tip
Integrate the orchestrator into existing CI/CD pipelines by using its gate status scripts to trigger automated tests and deployment checks, ensuring seamless alignment with tools like Jenkins or GitHub Actions for continuous validation.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Provides a 7-step debugging protocol plus language-specific commands to systematically identify, verify, and fix software bugs across multiple environments.
A comprehensive skill for using the Cursor CLI agent for various software engineering tasks (updated for 2026 features, includes tmux automation guide).
Write, run, and manage unit, integration, and E2E tests across TypeScript, Python, and Swift using recommended frameworks.
Control and operate Opencode via slash commands. Use this skill to manage sessions, select models, switch agents (plan/build), and coordinate coding through Opencode.
Coding style memory that adapts to your preferences, conventions, and patterns for consistent coding.