run-pipeline
Skill1 starsChanged 4 months agoMITView on GitHub
Adopt and execute the installed agents_pipeline full-pipeline workflow, including risk-derived planning, task routing, implementation, verification, review, retries, status artifacts, and cleanup. Use when the user explicitly invokes `$run-pipeline`, asks to use or run the agents_pipeline pipeline, or requests the full high-risk or multi-module workflow instead of a simple or bounded flow.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- harness/harness-skillstheir own version, first seen Nov 1, 2026
- neurospin-projects/2026_bdrabczuk_champollion_mcptheir own version, first seen Nov 1, 2026
- rgoerwit/ai-investment-agenttheir own version, first seen Oct 8, 2026
- ai-analyst-lab/ai-analysttheir own version, first seen Sep 10, 2026
- Duckz289/Pipeline-to-proposal-Plugin-STANDARDtheir own version, first seen Jul 27, 2026
- AlanSyue/my-skillstheir own version, first seen Feb 26, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- ai-analyst-lab/ai-analyst★ 304same name
- harness/harness-skills★ 127same name
- rgoerwit/ai-investment-agent★ 71same name
- AlanSyue/my-skills★ 2same name
- Duckz289/Pipeline-to-proposal-Plugin-STANDARD★ 2same name
- neurospin-projects/2026_bdrabczuk_champollion_mcp★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- artgen-scaffoldSkill
- decideSkill
- devtools-ux-auditSkill
- frontend-aesthetic-directorSkill
- run-adaptiveSkill
- run-analysisSkill
- run-ciSkill
- run-committeeSkill
- run-flowSkill
- run-generalSkill
- run-modernizeSkill
- run-simpleSkill
- run-specSkill
- run-uxSkill
- ui-communication-designerSkill
- ui-ux-workflowSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Read ${CODEX_HOME:-$HOME/.codex}/agents/orchestrator-pipeline.toml as the authoritative workflow definition.
- Very specific: line 15
If the global definition does not exist, stop and ask the user to run the global agents_pipeline Codex installer.
- Very specific: line 16
Treat its developer_instructions as authoritative for decomposition, delegation, artifacts, review, retries, cleanup, and final reporting.
- Distinctive: line 18
After the preflight succeeds, workspace profiles change effective role/model routing for dispatched agents.
- Very specific: line 22
Remove only the explicit $run-pipeline invocation token from the request.
- Distinctive: line 26
Keep status, checkpoint, resource teardown, evidence, retry, and resume behavior exactly as required by the installed definition.
“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Run Pipeline
Use the installed orchestrator-pipeline definition as the authoritative workflow. Keep actual role-model/tier selection in the effective Codex runtime/profile configuration and use the formal shared resolver for child-spawn effort only. Reasoning defaults to --reasoning=inherit; adaptive is the explicit selector-enforcement mode, while shadow computes without applying. --review=max remains an exact deep reviewer-only effort override, not certification or model routing.
Resolve the definition
- Resolve the current workspace root.
- Read
${CODEX_HOME:-$HOME/.codex}/agents/orchestrator-pipeline.tomlas the authoritative workflow definition. Never manually adopt a repository.codex/agents/orchestrator-pipeline.toml; Codex project trust and effective config, not raw file presence, control workspace role routing. - Always query the installed global profile manager for current-workspace JSON status before adopting the workflow. A normal workspace without a profile reports global inheritance and may continue. If status cannot be verified or a configured profile's
healthis notok, stop and ask the user to rerun workspacesetorclear; never dispatch through an unhealthy or orphaned profile. If a configured profile'scatalog_stateis notcurrent, stop before dispatch and require explicit workspaceset --model-set openaiorclear. If itsprofile_eligibilityis noteligible, warn that Codex is ignoring the workspace layer and continue with global role routing. Only a healthy, eligible, current profile may route dispatched roles locally. For that eligible path, retain each role's exact savedresolved_configurationfrom the selected versioned model set and registered reasoning projection, including its configuration identity, version, and digest. Pass that same envelope to the formal shared resolver and trace expectations. The resolver selects effort only, so native managed normal dispatch omits a raw model; a Web session agent Job selects only the saved role model underprotocols/REASONING_POLICY.md. Retired Codex model sets are not eligible for new dispatch. - If the global definition does not exist, stop and ask the user to run the global agents_pipeline Codex installer. Do not reconstruct the pipeline from memory.
- Treat its
developer_instructionsas authoritative for decomposition, delegation, artifacts, review, retries, cleanup, and final reporting.
After the preflight succeeds, workspace profiles change effective role/model routing for dispatched agents; they do not replace this global workflow source inside the current/main agent.
Execute the workflow
- Remove only the explicit
$run-pipelineinvocation token from the request. Preserve the remaining task text and pipeline flags as raw input. - Adopt the pipeline definition in the current/main agent. Do not spawn
orchestrator-pipelinemerely to enter the mode. - Obey the definition's hard constraints and delegation rules. If it routes scouting, implementation, testing, or review to helper roles, use those roles rather than bypassing them inline.
- Resolve each dispatched role through
protocols/REASONING_POLICY.mdwithout changing the current/main agent. Native managed dispatch uses effective workspace/global Codex role routing, omits full-history forks and normally omitsmodel; it verifies the local child trace before claiming adaptive enforcement. A Web session agent Job instead explicitly selects the saved role model, applies effort only in adaptive mode, observes one Job to terminal, and follows the policy's result and status limits without claiming native role identity. The sole native managed auto-recovery exception remains inprotocols/CAPABILITY_RECOVERY.mdand requires matching model/effort trace evidence.--review=maxremains an exact reviewer-only effort gate: adaptive requests and verifies it on the native surface, shadow records it without applying it, and inherit conflicts. It remains deep ordinary review, never applies to a non-review role, and is unavailable on the Job surface until supported evidence exists. Do not rewrite role files. - Keep status, checkpoint, resource teardown, evidence, retry, and resume behavior exactly as required by the installed definition.
- Return the definition's required user-facing synthesis, including changed paths, verification evidence, reviewer outcome, and explicit warnings or blockers.
Higher-priority runtime instructions and user approvals continue to apply.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

