agentsclimarketplace

Workflow launchpad

Skill Arnie016/codex-prompt-templates/skills/workflow-launchpad

Set up a practical Codex workflow for a new day, project, or repo. Use when the user wants an amazing workflow setup, a repeatable launch checklist, aligned skills, project memory, low-token defaults, daemon/process control, multi-agent coordination, or a personal Codex operating rhythm.From its SKILL.md

Install
npx -y skills add Arnie016/codex-prompt-templates --skill workflow-launchpad

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.5 KB, 501 tokens by cl100k_base, as published. Nobody here has run it

Workflow Launchpad

Use this to start work cleanly and avoid context drift.

Project Bootstrap

  1. Run $repo-intelligence-brief if the repo is unfamiliar.
  2. Run $low-token-context to create a compact task brief.
  3. Create .codex-memory/decisions.md only if durable decisions are expected.
  4. Choose one active workflow:
    • build
    • debug
    • review
    • research
    • experiment
    • MCP integration
  5. Define verification before editing.

Control Plane Setup

Use this when the project has daemons, background workers, queues, MCP/tool servers, release tasks, API-to-tool bridges, or parallel agents.

## Control Plane
Command surface:
Runtime state:
Lifecycle owner:
Ports/namespaces:
Status/logs:
Workers/queues:
Agent scopes/leases:
Trust boundaries:
Verification:
Stop/recover:
  • Write down the status/log/health command before starting processes.
  • Keep runtime state, credentials, caches, and local databases out of source control.
  • Assign write scopes or leases before parallel agent work.
  • Run $mcp-safety-review before trusting side-effecting tool surfaces.
  • Escalate production traces, evals, guardrails, budgets, shadow traffic, or release gates to $auto-skill-build-agent-reliability-loop.

Daily Rhythm

# Today
## Goal
## Context
## Active Files
## Commands
## Risks
## Done Criteria

End-Of-Session

  • Summarize decisions into memory.
  • Record failed attempts only if they prevent repeated work.
  • Update task brief with next action.
  • Keep logs out of memory unless compacted.

Use $codex-supercharge-router when the next workflow is unclear.

Read plugins/codex-supercharge/references/workflow-control-plane.md for the source-backed control-plane pattern.

Skip When

  • The task already has a precise plan and validation command.
  • The user only wants an immediate one-off answer.

Validation

  • The active workflow, verification command, and done criteria are written down.
  • Any durable decisions are placed in memory, not left only in chat.
  • The chosen skill path matches the current repo/task risk.
  • Daemon, worker, queue, MCP/tool, or multi-agent work has an explicit control plane before execution.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,614. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.