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 oil-oil/codex-team-mode --skill team-modegit clone --depth 1 https://github.com/oil-oil/codex-team-modeWrote 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/oil-oil/codex-team-mode/team-mode)<a href="https://agentmods.dev/skills/oil-oil/codex-team-mode/team-mode"><img src="https://agentmods.dev/badge/skills/oil-oil/codex-team-mode/team-mode/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/oil-oil/codex-team-mode/team-mode"><img src="https://agentmods.dev/badge/skills/oil-oil/codex-team-mode/team-mode.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00076 | $0.02748 |
| Opus 5 | $0.00038 | $0.01374 |
| Sonnet 5 | $0.00015 | $0.00550 |
| Haiku 4.5 | $0.00008 | $0.00275 |
Grade A, and why
team-mode 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 9d ago.
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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Team Mode
Lead the task in the main thread. Use the smallest useful parallel set of subagents to isolate noisy work, reduce expensive main-agent context, run independent work in parallel, or obtain a fresh review. Do not turn the roles into a mandatory pipeline.
Dispatch Gate
Every spawn_agent call must explicitly pass agent_type as exactly one of Explorer, Executor, or Reviewer.
- Never omit
agent_typeand never passdefault. Thedefaultprofile is a fail-closed dispatch guard, not a working Agent. - Never use
task_nameto select a profile; it only labels the child thread. - If the intended custom profile is unavailable, keep the work in the main thread or repair the profiles. Do not silently use a generic child.
- If a child returns the dispatch-guard message or its trace shows
default/subagent/unknown, reject its output and respawn only after selecting the intended custom profile explicitly.
Read references/custom-agents.md only when installing, repairing, or verifying the three working profiles and the default guard. The controlled onboarding self-test described there is the only time Team Mode deliberately omits agent_type.
One-Time Onboarding
Do not inspect Agent files, load onboarding instructions, or repeat setup explanations during normal Team Mode routing. Treat the active spawn_agent.agent_type choices and descriptions as runtime readiness evidence. When all three working profiles are available and default is described as the dispatch guard, skip onboarding without mentioning it.
Only read references/custom-agents.md and start onboarding when a required profile or guard is unavailable, or when the user explicitly asks to install, repair, verify, move, disable, or customize them. Get authorization before writing personal or project configuration. After a successful setup or repair, tell the user once what was installed, which models are assigned, that the guard affects omitted/default subagent dispatches in its installation scope, when restart or a new task is required, and how to disable or restore only the guard without removing the three working profiles.
What ships with it
5 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.
- 9d ago First seen · 121 lines · 76 tokens per session scan A 5a1f10c8801c
team-mode is a skill published in the GitHub repository oil-oil/codex-team-mode (118 stars, last pushed 1mo ago), licensed MIT. It adds 76 tokens to every session and 2,748 once invoked, about $0.0004 per session on Opus 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
workflow-design-bible
Generate a complete "constitution + documentation system" for a new autonomous, agent-run project — a content channel, an ebook press, an SEO tool-site, a web product, a casual game, or anything that should run itself with minimal human babysitting. Use when starting a brand-new self-running project and you want a…
mysql-mcp
Core guidelines and instruction pointers for the mysql-mcp server. Use when interacting with the mysql-mcp server, configuring tool filtering, using Code Mode (mysqlexecutecode), setting up database infrastructure, configuring observability, recovering from failures, or when you need to review its tools and…
datadog
Use when working with Datadog: monitoring, dashboards, APM, logs, RUM, SLOs, downtimes, or auth via the pup CLI. Triggers on: "Datadog", "DDAPIKEY", "pup", "monitors", "APM", "RUM", "SLO", "downtime". Do NOT trigger for generic observability/tracing without Datadog explicitly mentioned (use opentelemetry instead).
proxysql
Production standards for ProxySQL, a high-performance MySQL proxy providing connection multiplexing, query routing, caching, and firewalling. Use when configuring, troubleshooting, or making architectural decisions about multiplexing, query caching, query routing rules, or firewall lists. Triggers on: "ProxySQL"…
mysql-mcp-infrastructure
Master guide for orchestrating, recreating, and verifying the local database test ecosystems and examples in the mysql-mcp repository. Use when the user asks to set up the local test infrastructure, spin up the examples, or troubleshoot database cluster setups.
mysql-router
Production standards for MySQL Router, the lightweight middleware for transparent routing between applications and MySQL InnoDB Cluster servers. Use when configuring, troubleshooting, or making architectural decisions about HA routing, connection pooling, or cluster integration. Triggers on: "MySQL Router"…