Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add shinpr/codex-workflows --skill subagents-orchestration-guidegit clone --depth 1 https://github.com/shinpr/codex-workflowsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/shinpr/codex-workflows/subagents-orchestration-guide)<a href="https://agentmods.dev/skills/shinpr/codex-workflows/subagents-orchestration-guide"><img src="https://agentmods.dev/badge/skills/shinpr/codex-workflows/subagents-orchestration-guide/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/shinpr/codex-workflows/subagents-orchestration-guide"><img src="https://agentmods.dev/badge/skills/shinpr/codex-workflows/subagents-orchestration-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Memory Poisoning · line 86 Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
- medium Excessive Agency · line 235 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00037 | $0.06241 |
| Opus 5.5 | $0.00015 | $0.02496 |
| Sonnet 5.5 | $0.00007 | $0.01248 |
| Haiku 4.5 | $0.00004 | $0.00624 |
Grade A, and why
subagents-orchestration-guide scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 348 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Subagents Orchestration Guide
Load subagent-delegation for assignment, waiting, and intervention rules. This guide owns workflow-specific coordination.
Invocation rule: Start each named agent with spawn_agent and fork_turns="none". When invoking the same named agent again within this flow, use followup_task on its existing instance with the inputs specified for the current invocation.
Role: The Orchestrator
The orchestrator owns workflow state and directly performs lightweight coordination: locating artifacts, reading status fields, resolving paths, applying explicit user answers, updating execution plans and continuation records, running deterministic repository checks, and filling specialist input contracts from their sources. Invoke a specialized subagent when a recipe names that role or the work requires its domain judgment or implementation authority.
Purpose and Decision Authority
Every role owns the smallest sufficient result within its assigned scope. Unnecessary mechanisms create downstream implementation, verification, and maintenance work. Apply ai-development-guide's Outcome Boundary: evidence and prior passage do not establish necessity or lifecycle value.
A phase pass means the current evidence permits progression. User confirmation permits the requested next work; neither makes internal design choices immutable. Final user acceptance concerns the delivered outcome. Interpret existing approved or ADR Accepted records as passage/continuation in context, without requiring a new field or repeating permission already granted.
Interpret user responses by their meaning and the authority they supply. Continuation or route approval advances the workflow. A product requirement or exclusion becomes settled when the user's response selects it.
Resolve outcome-preserving technical reductions, including superseding ADR choices, through existing owners. Update only affected sources, tasks, and proof, and continue within the user's execution authority. Escalate changes to user-required outcomes, explicit constraints, actual consumer obligations, or external-action authority. Optional new requirements remain unadopted; propose a requirement change only with evidence that the current conditions cannot deliver the requested result.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed fcd097b801a7
- 6d ago Changed 863e939c0a60
- 26d ago Changed e876287c0d3c
- 27d ago Changed · +5 lines 402a82c9b0de
- 1mo ago Changed · +8 lines f25d9edb8005
- 1mo ago Changed · +1 lines 9131c4d86fcc
- 1mo ago Changed · +2 lines ae078fa6840b
- 1mo ago Changed · -2 lines 9244b0b55b5b
- 1mo ago Changed · +10 lines ac3164a2fcd7
- 1mo ago First seen · 324 lines · 37 tokens per session scan A cfe4b5578a83
subagents-orchestration-guide is a skill published in the GitHub repository shinpr/codex-workflows (39 stars, last pushed today), licensed MIT. It adds 37 tokens to every session and 6,241 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
chrome-cdp
Drive a headless Chrome over the Chrome DevTools Protocol (CDP) for browser QA — navigate, click, fill forms, read the DOM/accessibility tree, screenshot, and assert. Use whenever a task requires loading a web page and interacting with it like a user. Chrome is launched by a bash step (recipe below); this skill…
skill-creator
Create, install, or update skills in the workspace. Use when (1) installing a skill from a URL or remote source, (2) creating a new skill from scratch, (3) updating or restructuring existing skills. Always use this skill for any skill installation or creation task.
new-app
Scaffold a new Atomic Agents project from scratch — create the directory, pyproject.toml, env file, first agent, and a runnable entry point. Use when the user asks to start a new atomic-agents project from scratch, says "scaffold" / "new project" / "start from zero", or runs /atomic-agents:new-app.
ax-agent-rlm
This skill helps an LLM generate correct AxAgent RLM/runtime code using @ax-llm/ax. Use when the user asks about RLM code execution, AxJSRuntime, contextFields, contextPolicy, liveRuntimeState, promptLevel, stage prompt controls, executorModelPolicy, maxRuntimeChars, agent.test(...), llmQuery(...), recursionOptions…
ax-go-flow
Use when writing Go code with github.com/ax-llm/ax/packages/go for flows, nodes, program graphs, nested programs, dynamic options, caching, and optimizer components.
notifications
Send notifications through the unified notification router.