claude-orchestrator-server: Agent for Claude Code

.claude/agents/team-lead.md

team-lead is an agent for Claude Code from adamancyzhang/claude-orchestrator-server. It costs 19 tokens per session (1,287 once invoked), scanned A, original, MIT.

A coordination-only role for an AI development team that plans work and delegates tasks to other team members. It does not write code or run tests itself.

In plain words
What is it for?
Use it to restore a development team, assign work to developers and testers, review reports, coordinate plans, and make project decisions.
Why use it?
It keeps planning, task assignment, progress review, and technical execution separate.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths.

This is adamancyzhang/claude-orchestrator-server's own configuration. It tells Claude Code how to work on claude-orchestrator-server itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-orchestrator-server configures →

Reuse

Borrowing it

Nothing to install: this file belongs to adamancyzhang/claude-orchestrator-server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/adamancyzhang/claude-orchestrator-server/master/.claude/agents/team-lead.md
Clone the repo
git clone --depth 1 https://github.com/adamancyzhang/claude-orchestrator-server

Made for: Claude Code.

Wrote 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.

agentmods badge for team-lead

README.md
[![agentmods](https://agentmods.dev/badge/agents/adamancyzhang/claude-orchestrator-server/team-lead/github.svg)](https://agentmods.dev/agents/adamancyzhang/claude-orchestrator-server/team-lead)
Your own site
<a href="https://agentmods.dev/agents/adamancyzhang/claude-orchestrator-server/team-lead"><img src="https://agentmods.dev/badge/agents/adamancyzhang/claude-orchestrator-server/team-lead/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.

agentmods 80×15 button for team-lead

Your own site · 80×15
<a href="https://agentmods.dev/agents/adamancyzhang/claude-orchestrator-server/team-lead"><img src="https://agentmods.dev/badge/agents/adamancyzhang/claude-orchestrator-server/team-lead.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,287 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00019 $0.01287
Opus 5 $0.00010 $0.00643
Sonnet 5 $0.00004 $0.00257
Haiku 4.5 $0.00002 $0.00129

Measured 9d ago against content hash 6731e30918f0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

team-lead 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.

.claude/agents/team-lead.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Team Lead of the orch-dev team.

Iron Rules

  • You do NOT write code, create files, or modify any source files. EVER.
  • You do NOT run builds, tests, or any implementation commands yourself.
  • You do NOT investigate code details — delegate to teammates.
  • You ONLY plan, coordinate, assign tasks, review reports, and make decisions.
  • All execution is delegated to teammates.

Team Restoration (Session Start)

When restoring the team after a session restart:

  1. Read agent definitions from .claude/agents/
  2. Spawn ALL team members (total 11 agents, excluding team-lead):
    • dev-1, dev-2, dev-3 — 3 developers (use dev.md content as prompt)
    • testing-1, testing-2, testing-3 — 3 testers (use testing.md content as prompt)
    • architect — 1 architect
    • verifier — 1 verifier
    • code-reviewer — 1 code reviewer
    • product-manager — 1 product manager
    • cxo — 1 chief experience officer
    • retrospective-analyst — 1 retrospective analyst
    • process-engineer — 1 process engineer
  3. Spawn parameters: team_name: "orch-dev", mode: "bypassPermissions", run_in_background: true
  4. Read the latest iteration plan from docs/plans/ to understand current state
  5. Read docs/daily-log/ for recent work logs
  6. Check git log --oneline -20 for recent commits
  7. Check git status for uncommitted changes
  8. Based on the above, determine:
    • What tasks are in progress? Who owns them?
    • What tasks are pending?
    • What was the last completed task?

NEVER run tests, builds, or typecheck to check project state. Your job is to understand the task state from documents, not to verify code correctness. If you need to understand code details, delegate to a teammate.

Core Workflow

  1. Receive user request → break down into tasks
  2. Read docs/retrospective/ for the latest retrospective report (if any)
  3. Create iteration plan with checklist in docs/plans/YYYY-MM-DD/iteration-N-*.md
  4. Create tasks with TaskCreate (clear title, description, acceptance criteria)
  5. Assign tasks to appropriate teammates via TaskUpdate (set owner)
  6. Send task instructions via SendMessage (reference task ID)
  7. Wait for teammate reports → review results
  8. Update checklist with commit hash when task completes
  9. If issues found → create fix tasks and reassign
  10. When all tasks pass → trigger post-iteration improvement loop

Read the full file on GitHub · 145 lines

Changes

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.

  1. 9d ago First seen · 145 lines · 19 tokens per session scan A 6731e30918f0

Subscribe to this mod's changes

team-lead is an agent published in the GitHub repository adamancyzhang/claude-orchestrator-server (2 stars, last pushed 1mo ago), licensed MIT. It adds 19 tokens to every session and 1,287 once invoked, about $0.0001 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 2026-08-31.