{ "schemaVersion": "1.0", "item": { "slug": "codex-conductor", "name": "Codex Conductor", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/ShalomObongo/codex-conductor", "canonicalUrl": "https://clawhub.ai/ShalomObongo/codex-conductor", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/codex-conductor", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=codex-conductor", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "references/codex-runbook.md", "references/gate-checklists.md", "references/gate-prompts.md", "references/manual-test-templates.md", "references/modes.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-23T16:43:11.935Z", "expiresAt": "2026-04-30T16:43:11.935Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/codex-conductor" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/codex-conductor", "agentPageUrl": "https://openagent3.xyz/skills/codex-conductor/agent", "manifestUrl": "https://openagent3.xyz/skills/codex-conductor/agent.json", "briefUrl": "https://openagent3.xyz/skills/codex-conductor/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "Codex Orchestrator", "body": "Coordinate Codex as a disciplined delivery system, not a one-shot generator." }, { "title": "Core Modes", "body": "Select both:\n\nproject_mode\n\ngreenfield: build from scratch\nbrownfield: onboard and modernize an existing system\n\n\nexecution_mode\n\nautonomous: proceed automatically when gates pass\ngated: pause at every gate for user approval" }, { "title": "Governing Principle: Spec-Driven Development", "body": "No code without a spec. This is non-negotiable.\n\nBefore any implementation, a written spec must exist with:\n\nWhat is being built\nWhy it's needed\nAcceptance criteria (testable)\nConstraints and out-of-scope\n\nThe coding agent MUST NOT:\n\nGuess at requirements\nMake assumptions about behavior\nAdd unrequested features\nInvent abstractions not in spec\n\nIf spec is unclear → STOP and ask. Never guess.\n\nSee references/spec-driven-development.md for full spec templates and enforcement rules." }, { "title": "Non-Negotiable Sequence", "body": "Intake + planning questionnaire\nSpec creation + approval (specs written BEFORE any code)\nDocs scaffold + AGENTS.md contract\nMode-specific pre-architecture work\nArchitecture + ADR baseline (references specs)\nBuild by vertical slices (each task references spec)\nVerification against spec acceptance criteria\nSecurity/quality gates\nRelease readiness + handover\n\nNever skip gates silently. Never implement without a spec." }, { "title": "Required Resources", "body": "Read these references before running:\n\nreferences/spec-driven-development.md (MANDATORY FIRST - governs all work)\nreferences/planning-questionnaire.md\nreferences/modes.md\nreferences/gate-checklists.md\nreferences/testing-matrix.md\nreferences/manual-test-templates.md\nreferences/codex-runbook.md\nreferences/gate-prompts.md\nscripts/agent_exec.py\nreferences/research-playbook.md (if research_mode=true)" }, { "title": "Scaffolding", "body": "Initialize project artifacts:\n\npython scripts/init_project_docs.py --root --mode \n\nThis creates/updates:\n\nAGENTS.md (project workflow contract)\ndocs/*.md planning/architecture/test/progress/change docs\nbrownfield docs (when mode is brownfield)\n.orchestrator/status.json (machine-readable state)\n.orchestrator/context.json (project/execution/research mode context)" }, { "title": "Planning Rules", "body": "Before anything else, ask the user which coding agent to use (codex | claude | opencode | pi) and fallback agent.\nThen ask all required questions from references/planning-questionnaire.md.\n\nMinimum required answers:\n\nmission\ntop user journeys\nv1 scope\nhosting target\nstack preference (or explicit request for recommendation)\nproject_mode\nexecution_mode\ndefinition of done\nacceptance tests\n\nIf research_mode=true, produce docs/research-notes.md and architecture recommendation before G2." }, { "title": "Greenfield", "body": "Must complete before G2:\n\nrequirements + DoD clarity\narchitecture baseline\nADR-0001 with alternatives\nCI/test baseline plan" }, { "title": "Brownfield", "body": "Must complete before G2 (and authored by coding agent, not orchestrator):\n\nas-is architecture and system inventory\ndependency map and risk register\ncharacterization-test baseline\nmigration strategy + rollback approach\ncompatibility boundaries documented" }, { "title": "Gate Engine", "body": "Use gates G0 through G7 defined in references/gate-checklists.md.\n\nUpdate gate state via script:\n\npython scripts/gate_status.py set --root --gate G3 --state PASS --note \"slice-1 verified\"\n\nValidate status schema:\n\npython scripts/gate_status.py validate --root \n\nAllowed states: PENDING | IN_PROGRESS | PASS | FAIL | BLOCKED.\nBy default, gate preconditions are enforced (sequence + mode-aware docs checks)." }, { "title": "Validation Rules", "body": "Use references/testing-matrix.md.\n\nMandatory checks per progression:\n\nlint/type/build\nunit/integration/e2e (as applicable)\nAPI contract sanity (if API exists)\nsecurity baseline\ndocs sync verification\n\nAlso execute manual test scripts from references/manual-test-templates.md." }, { "title": "Documentation Rules", "body": "For each meaningful step:\n\nupdate docs/tasks.md\nupdate docs/progress.md\nappend docs/change-log.md\nupdate docs/traceability.md\nrecord test evidence in docs/test-results.md\n\nFor user-requested changes, run:\n\npython scripts/change_impact.py --root --request \"\"\n\nThen complete all TODOs it emits in impacted docs." }, { "title": "Codex Execution Pattern", "body": "Use PTY/background for long runs. Follow command patterns in references/codex-runbook.md.\n\nCritical rule: each run executes ONE task, not a whole project in one prompt.\nFor G4, maintain docs/g4-task-plan.md checklist and process tasks one by one.\n\nGenerate gate-specific prompts with:\n\npython scripts/generate_gate_prompt.py --gate --agent --project-mode --execution-mode --research-mode --task \"\" --spec-ref \"\"\n\nupdate_docs_step.py is now a fallback utility for recovery/manual bookkeeping only.\nPrimary expectation: the coding agent updates docs directly during each task.\n\nRequired loop:\n\nverify spec exists for the task (no spec = no implementation)\nlaunch selected coding agent with spec-driven prompt template\ncoding agent updates docs immediately after task completion (including handoff checklist)\ncoding agent wakes OpenClaw with task summary + where verification steps are documented\nOpenClaw agent runs verification itself:\n\nCLI checks in terminal tools\nBrowser/manual checks in browser tools (for web flows)\n\n\nverify output matches spec acceptance criteria\nif validations fail, OpenClaw sends exact failures back to coding agent and re-runs fix cycle\nwrite final gate status only after validations pass (or mark FAIL/BLOCKED)\n\nEnforcement:\n\nrun_gate.py requires --spec-ref for G3/G4 tasks (implementation gates).\nrun_gate.py requires coding agent + fallback agent context.\nEach task requires validation evidence (--validate-cmd and/or --ui-review-note).\nTasks flagged with --requires-browser-check must include --ui-review-note.\nstatus=PASS requires at least one --validate-cmd.\nstatus=PASS is blocked when --agent-dry-run is used.\nFor G4, PASS is blocked until docs/g4-task-plan.md has no unchecked tasks.\nValidation output is recorded in docs/validation-log.md.\nCoding agent must update docs after every task, including docs/agent-handoff.md.\nIn brownfield mode, G1/G2 fail if onboarding docs are not updated by the coding agent.\nCoding agent prompts MUST include spec preamble from references/spec-driven-development.md.\nAny implementation without spec reference = automatic FAIL.\nIn autonomous mode, failed validations trigger automatic fix retries (default: 2) with failure details passed back to coding agent.\nOptional strict mode: --auto-block-on-retry-exhaust auto-classifies gate as BLOCKED when retries are exhausted." }, { "title": "Progress Visibility", "body": "Generate a quick status board:\n\npython scripts/progress_dashboard.py --root \n\nThis summarizes current gate, completion %, blockers, and recent activity.\n\nRun a single-task gate step with one command:\n\npython scripts/run_gate.py --root --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\"\n\nMark PASS only after all gate-level checklist items are complete:\n\npython scripts/run_gate.py --root --gate G2 --agent codex --task \"architecture gate complete\" --status PASS --validate-cmd \"npm run -s typecheck\"\n\nFor web/UI tasks, require browser verification by OpenClaw agent:\n\npython scripts/run_gate.py ... --requires-browser-check --ui-review-note \"Verified login + CRUD manually in browser via OpenClaw browser tools\"\n\nPackage distributable skill artifact:\n\npython scripts/package_skill.py --skill-dir . --out dist" }, { "title": "End-State Deliverables", "body": "At completion provide:\n\ndocs/progress.md at 100%\nfinal gate summary from .orchestrator/status.json\ntest result summary + unresolved risks\ndeployment + rollback notes\nnext-iteration backlog\n\nIf blockers remain, mark as PARTIAL_COMPLETE with explicit blockers and owners." } ], "body": "Codex Orchestrator\n\nCoordinate Codex as a disciplined delivery system, not a one-shot generator.\n\nCore Modes\n\nSelect both:\n\nproject_mode\ngreenfield: build from scratch\nbrownfield: onboard and modernize an existing system\nexecution_mode\nautonomous: proceed automatically when gates pass\ngated: pause at every gate for user approval\nGoverning Principle: Spec-Driven Development\n\nNo code without a spec. This is non-negotiable.\n\nBefore any implementation, a written spec must exist with:\n\nWhat is being built\nWhy it's needed\nAcceptance criteria (testable)\nConstraints and out-of-scope\n\nThe coding agent MUST NOT:\n\nGuess at requirements\nMake assumptions about behavior\nAdd unrequested features\nInvent abstractions not in spec\n\nIf spec is unclear → STOP and ask. Never guess.\n\nSee references/spec-driven-development.md for full spec templates and enforcement rules.\n\nNon-Negotiable Sequence\nIntake + planning questionnaire\nSpec creation + approval (specs written BEFORE any code)\nDocs scaffold + AGENTS.md contract\nMode-specific pre-architecture work\nArchitecture + ADR baseline (references specs)\nBuild by vertical slices (each task references spec)\nVerification against spec acceptance criteria\nSecurity/quality gates\nRelease readiness + handover\n\nNever skip gates silently. Never implement without a spec.\n\nRequired Resources\n\nRead these references before running:\n\nreferences/spec-driven-development.md (MANDATORY FIRST - governs all work)\nreferences/planning-questionnaire.md\nreferences/modes.md\nreferences/gate-checklists.md\nreferences/testing-matrix.md\nreferences/manual-test-templates.md\nreferences/codex-runbook.md\nreferences/gate-prompts.md\nscripts/agent_exec.py\nreferences/research-playbook.md (if research_mode=true)\nScaffolding\n\nInitialize project artifacts:\n\npython scripts/init_project_docs.py --root --mode \n\n\nThis creates/updates:\n\nAGENTS.md (project workflow contract)\ndocs/*.md planning/architecture/test/progress/change docs\nbrownfield docs (when mode is brownfield)\n.orchestrator/status.json (machine-readable state)\n.orchestrator/context.json (project/execution/research mode context)\nPlanning Rules\n\nBefore 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.\n\nMinimum required answers:\n\nmission\ntop user journeys\nv1 scope\nhosting target\nstack preference (or explicit request for recommendation)\nproject_mode\nexecution_mode\ndefinition of done\nacceptance tests\n\nIf research_mode=true, produce docs/research-notes.md and architecture recommendation before G2.\n\nMode-Specific Requirements\nGreenfield\n\nMust complete before G2:\n\nrequirements + DoD clarity\narchitecture baseline\nADR-0001 with alternatives\nCI/test baseline plan\nBrownfield\n\nMust complete before G2 (and authored by coding agent, not orchestrator):\n\nas-is architecture and system inventory\ndependency map and risk register\ncharacterization-test baseline\nmigration strategy + rollback approach\ncompatibility boundaries documented\nGate Engine\n\nUse gates G0 through G7 defined in references/gate-checklists.md.\n\nUpdate gate state via script:\n\npython scripts/gate_status.py set --root --gate G3 --state PASS --note \"slice-1 verified\"\n\n\nValidate status schema:\n\npython scripts/gate_status.py validate --root \n\n\nAllowed states: PENDING | IN_PROGRESS | PASS | FAIL | BLOCKED. By default, gate preconditions are enforced (sequence + mode-aware docs checks).\n\nValidation Rules\n\nUse references/testing-matrix.md.\n\nMandatory checks per progression:\n\nlint/type/build\nunit/integration/e2e (as applicable)\nAPI contract sanity (if API exists)\nsecurity baseline\ndocs sync verification\n\nAlso execute manual test scripts from references/manual-test-templates.md.\n\nDocumentation Rules\n\nFor each meaningful step:\n\nupdate docs/tasks.md\nupdate docs/progress.md\nappend docs/change-log.md\nupdate docs/traceability.md\nrecord test evidence in docs/test-results.md\n\nFor user-requested changes, run:\n\npython scripts/change_impact.py --root --request \"\"\n\n\nThen complete all TODOs it emits in impacted docs.\n\nCodex Execution Pattern\n\nUse PTY/background for long runs. Follow command patterns in references/codex-runbook.md.\n\nCritical 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.\n\nGenerate gate-specific prompts with:\n\npython scripts/generate_gate_prompt.py --gate --agent --project-mode --execution-mode --research-mode --task \"\" --spec-ref \"\"\n\n\nupdate_docs_step.py is now a fallback utility for recovery/manual bookkeeping only. Primary expectation: the coding agent updates docs directly during each task.\n\nRequired loop:\n\nverify spec exists for the task (no spec = no implementation)\nlaunch selected coding agent with spec-driven prompt template\ncoding agent updates docs immediately after task completion (including handoff checklist)\ncoding agent wakes OpenClaw with task summary + where verification steps are documented\nOpenClaw agent runs verification itself:\nCLI checks in terminal tools\nBrowser/manual checks in browser tools (for web flows)\nverify output matches spec acceptance criteria\nif validations fail, OpenClaw sends exact failures back to coding agent and re-runs fix cycle\nwrite final gate status only after validations pass (or mark FAIL/BLOCKED)\n\nEnforcement:\n\nrun_gate.py requires --spec-ref for G3/G4 tasks (implementation gates).\nrun_gate.py requires coding agent + fallback agent context.\nEach task requires validation evidence (--validate-cmd and/or --ui-review-note).\nTasks flagged with --requires-browser-check must include --ui-review-note.\nstatus=PASS requires at least one --validate-cmd.\nstatus=PASS is blocked when --agent-dry-run is used.\nFor G4, PASS is blocked until docs/g4-task-plan.md has no unchecked tasks.\nValidation output is recorded in docs/validation-log.md.\nCoding agent must update docs after every task, including docs/agent-handoff.md.\nIn brownfield mode, G1/G2 fail if onboarding docs are not updated by the coding agent.\nCoding agent prompts MUST include spec preamble from references/spec-driven-development.md.\nAny implementation without spec reference = automatic FAIL.\nIn autonomous mode, failed validations trigger automatic fix retries (default: 2) with failure details passed back to coding agent.\nOptional strict mode: --auto-block-on-retry-exhaust auto-classifies gate as BLOCKED when retries are exhausted.\nProgress Visibility\n\nGenerate a quick status board:\n\npython scripts/progress_dashboard.py --root \n\n\nThis summarizes current gate, completion %, blockers, and recent activity.\n\nRun a single-task gate step with one command:\n\npython scripts/run_gate.py --root --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\"\n\n\nMark PASS only after all gate-level checklist items are complete:\n\npython scripts/run_gate.py --root --gate G2 --agent codex --task \"architecture gate complete\" --status PASS --validate-cmd \"npm run -s typecheck\"\n\n\nFor web/UI tasks, require browser verification by OpenClaw agent:\n\npython scripts/run_gate.py ... --requires-browser-check --ui-review-note \"Verified login + CRUD manually in browser via OpenClaw browser tools\"\n\n\nPackage distributable skill artifact:\n\npython scripts/package_skill.py --skill-dir . --out dist\n\nEnd-State Deliverables\n\nAt completion provide:\n\ndocs/progress.md at 100%\nfinal gate summary from .orchestrator/status.json\ntest result summary + unresolved risks\ndeployment + rollback notes\nnext-iteration backlog\n\nIf blockers remain, mark as PARTIAL_COMPLETE with explicit blockers and owners." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/ShalomObongo/codex-conductor", "publisherUrl": "https://clawhub.ai/ShalomObongo/codex-conductor", "owner": "ShalomObongo", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/codex-conductor", "downloadUrl": "https://openagent3.xyz/downloads/codex-conductor", "agentUrl": "https://openagent3.xyz/skills/codex-conductor/agent", "manifestUrl": "https://openagent3.xyz/skills/codex-conductor/agent.json", "briefUrl": "https://openagent3.xyz/skills/codex-conductor/agent.md" } }