Subagent Roster Designer for Your Coding Project

Produces a 3-5 subagent roster with system prompts, minimal tool permissions, model tiers and explicit handoff rules for your project.

You get: Roster table, A ready-to-paste system prompt under 120 words for each… and handoff rules as 'From -> To

More like this

Ready to use: no blanks to fill in.

The prompt

subagent-roster-designer-for-your-coding-project.md
Act as a senior engineer who designs agent teams for software projects. My job is to decide which 3-5 subagents to set up for my project and how they pass work to each other. If I have not told you the project type, main language or stack, and which tools my agent platform allows, ask me up to two short questions first, then wait for my answer.

Rules:
- Give each subagent one narrow responsibility with no overlap.
- Grant the minimum tools needed, such as read-only access for reviewers.
- Choose a model tier per agent (stronger for planning and review, cheaper for repetitive work) with a one-line reason.
- Handoff rules must name the exact artifact passed on and when the agent must stop and ask me.

Output format:
1. Assumptions (separate list).
2. Roster table: agent, purpose, allowed tools, model tier, reason.
3. A ready-to-paste system prompt under 120 words for each agent.
4. Handoff rules as 'From -> To: artifact, done-when, failure action'.
5. Risks and missing information.
6. Three next actions.

Before answering, check that no two agents share a job, every tool is justified, and I could set this up today.

Example output

AI-written example

1. Assumptions

  • Next.js 14 + TypeScript app, Prisma/Postgres, Jest and Playwright, GitHub PRs.
  • The platform allows per-agent tool lists: Read, Grep, Glob, Edit, Write, Bash.
  • One developer; the main session acts as orchestrator.

2. Roster

AgentPurposeToolsTierReason
plannerTicket to task specRead, Grep, GlobStrongBad plans cost the most
implementerCode the specRead, Grep, Glob, Edit, Write, BashMidWell-scoped coding
testerWrite and run testsRead, Grep, Glob, Edit, BashCheapRepetitive loop
reviewerAudit the diffRead, Grep, Glob, Bash (git only)StrongLast line of defense

3. System prompts

  • planner: You write specs, never code. Read the ticket and codebase, then produce spec.md: files to change, acceptance criteria, out-of-scope items. If the ticket is ambiguous or touches auth or the DB schema, stop and ask the user.
  • implementer: Implement only what spec.md states. Touch no test files. Run typecheck and lint before finishing. After two failed attempts at the same problem, stop and ask. Output a short change-notes.md.
  • tester: Edit only files under __tests__/ and e2e/. Write tests for each acceptance criterion, run the suite, and report results in test-report.md. Never modify source code to make tests pass.
  • reviewer: You are read-only. Run git diff, check for bugs, security issues, and spec deviations. Write review.md listing blockers and nits with file and line. Never fix anything yourself.

4. Handoff rules

  • Me -> planner: ticket text. Done when spec.md has testable cri
Generated from this prompt by AI. Your results will vary.

Tips and common edits

Use when you are about to configure subagents for a real codebase. Tell the model your project type, stack and available tools, or answer its two questions.

Questions about this prompt

Parallel Research Fan-Out and Fan-In with Citation Table and Decision Memo

Splits a research question across parallel research agents, defines source discovery and evidence review steps, merges findings into a citation table, and produces a decision memo without claiming live browsing.

Research-to-Report Pipeline with Planner, Researcher, Writer and Fact-Checker

Designs a four-agent pipeline for your report type with each agent's prompt, inputs, outputs and handoff format.

Strict JSON Handoff Schemas with Validation and Retry Rules for Agent Pipelines

Turns a pasted pipeline description into JSON schemas for every agent-to-agent handoff, plus validation checks and retry rules for bad output.

Agent Pipeline Failure Mode Audit with Retry, Fallback, and Human Escalation Rules

Audits an existing multi-agent pipeline for failure modes and produces retry limits, fallback paths, escalation triggers, and a test plan. Supports the decision on what to harden before wider rollout.

Sequential Agent Handoff Contract Builder for Research-to-Publish Chains

Designs the handoff contracts, stop conditions, and review gates for a sequential chain such as researcher, writer, editor, and fact-checker. Helps you decide whether the chain is ready to run.

20-Case Agent Eval Set with Pass/Fail Rubric for Normal, Edge and Unsafe Requests

Builds 20 test cases for your agent's task, with expected behavior, failure signals and a pass/fail rubric covering normal, edge and unsafe requests.

Agent Tool Names, Descriptions and JSON Schemas from API Docs

Turns pasted API docs or a function list into agent-ready tool names, descriptions, JSON input schemas and when-to-use and when-not-to-use notes.

Agent Workflow Decomposition with Autonomy Levels and Stop Rules

Turns a manual process into a step-by-step agent plan with tools, approval gates, retry limits, and a build/pilot/skip recommendation.

Characterization Tests Before a Refactor

Writes tests that lock in a module's current behavior, including edge cases and quirks, so you can refactor safely.

Checkable Guardrail Rules for a Coding Agent: Blocked Commands, Approvals, Secrets and Hooks

Produces a set of testable guardrail rules for your coding agent, covering blocked commands, approval gates, secret handling and hook checks.

Codebase Tour for a Specific Task

Explores a repo through the lens of your task and explains the architecture, key files and data flow, then says where to start.

Conventional Commit and PR Description from a Diff and Ticket

Produces a Conventional Commits message and a reviewer-ready PR description covering what changed, why, how to test it, and the risks, based on your diff and ticket.