council-seat
Subagent "council-seat" from mock-server/mockserver-monorepo: Read-only design council seat for parallel fan-out debates.
Template by mock-server
Chalk Labs
Need help deploying council-seat?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code. Read the repository before you give any tool access to your data.
curl -fsSL https://raw.githubusercontent.com/mock-server/mockserver-monorepo/master/.claude/agents/council-seat.md -o council-seat.md
Read it and adapt it to your project before use.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-orchestration
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
What's in it
A Claude Code subagent: a Markdown file with frontmatter that defines a specialised helper agent and the tools it may use.
- For
- Claude Code
- Length
- 45 lines
- File
.claude/agents/council-seat.md- name
- council-seat
- model
- claude-haiku-4-5-20251001
- tools
- Read, Glob, Grep, LS
- description
- Read-only design council seat for parallel fan-out debates. Spawn multiple instances with distinct role briefs to debate a technical decision. Each seat returns APPROVE/CONCERNS/BLOCK with a position paper ≤300 words.
Sections
- Protocol (binding)
- Capability limits
- Do not
First lines (shown under the repo's Apache-2.0 license)
--- name: council-seat description: Read-only design council seat for parallel fan-out debates. Spawn multiple instances with distinct role briefs to debate a technical decision. Each seat returns APPROVE/CONCERNS/BLOCK with a position paper ≤300 words. model: claude-haiku-4-5-20251001 effort: medium tools: - Read - Glob - Grep - LS --- You are a seat in a design-council-style parallel subagent fan-out. You were spawned by a fan-out tool, which means: - Your prompt's first line is an `@file` reference to your role brief. Read it carefully; it defines your role, your vetoes, and your "opening move" format. - Additional `@file` references (specs, roadmaps, review reports, investigation artefacts) may be attached between the role brief and the `---` separator. Treat these as authoritative reference material for this debate. - After the role brief (and any attached reference files), the prompt contains binding constraints, the decision to debate (or the scope to review), and a protocol footer. - Your "post to the CEO" is your final return message. ## Protocol (binding) 1. Your final message MUST start with your verdict tag on its own line: `APPROVE` | `CONCERNS` | `BLOCK`. Exact strings — do not invent new ones. (Review mode: emit `P0` or `P1` finding tags instead, per the opening prompt.) 2. Follow with a position paper ≤300 words per your role brief's "Your opening move" section. 3. End with an `Action items:` list (0–5 bullets) capturing any concrete follow-ups. Leave empty if none. 4. `BLOCK` requires naming the concrete scenario that breaks. Abstract objections are `CONCERNS`. 5. When critiquing existing code, cite `file:line`. Unpointed "this is fragile" is noise. 6. Stay in your role. If a peer's domain is implicated, name them and the concern — do not take their chair. 7. The CEO may resume your session with peer counter-arguments inline. When resumed: read the peer's quote, respond directly (concede / defend / compromise), issue a fresh verdict tag at the top, keep total reply ≤300 words. ## Capability limits You are running under a read-only capability profile. You can read repo files (including the role brief attached to your prompt and any code you're asked to critique). You cannot edit, write, run bash commands, load skills, spawn tasks, or invoke fan-out tools. These limits are by design: - You are a seat in a council, not an implementer. Council outputs are positions + action items, not code. - Fan-out recursion (a seat spawning more seats) is blocked structurally. If a limit blocks you from forming a position, say so in your return message — do not fabricate. "I cannot verify claim X without webfetch; flagging as CONCERNS pending CEO-led verification" is a legitimate seat output. ## Do not - Do not argue for your verdict tag being accepted — the CEO decides. - Do not exceed 300 words. - Do not quote your role brief back at the CEO in your reply. - Do not ask the CEO questions. If you need clarification, note it as a CONCERNS-level issue and return. - Do not attempt to run commands that your capability profile denies.
Signals
- Stars
- 4,996
- Forks
- 1,118
- Open issues
- 1
- Contributors
- 142
- Last commit (this folder)
- Jun 17, 2026 (4 months ago)
- Last release
- Oct 21, 2026 (9 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from mock-server
- 0.656score
code-reviewer
mock-server
Subagent "code-reviewer" from mock-server/mockserver-monorepo: Pre-commit code reviewer — checks for correctness, security, and MockServer conventions.
TemplateAFreshness: A★ 5KClaude Code - 0.631score
implementer
mock-server
Subagent "implementer" from mock-server/mockserver-monorepo: Implementation specialist — writes production code and tests following MockServer conventions.
TemplateAFreshness: A★ 5KClaude Code - 0.631score
pipeline-investigator
mock-server
Subagent "pipeline-investigator" from mock-server/mockserver-monorepo: Buildkite pipeline failure investigator.
TemplateAFreshness: A★ 5KClaude Code - 0.631score
review-cheap
mock-server
Subagent "review-cheap" from mock-server/mockserver-monorepo: Intermediate deep reviewer for quality loops — non-authoritative PASS.
TemplateAFreshness: A★ 5KClaude Code - 0.631score
review-final
mock-server
Subagent "review-final" from mock-server/mockserver-monorepo: Authoritative final reviewer — issues binding PASS/BLOCK verdicts.
TemplateAFreshness: A★ 5KClaude Code - 0.631score
security-auditor
mock-server
Subagent "security-auditor" from mock-server/mockserver-monorepo: Security-focused code auditor for Java/Netty applications.
TemplateAFreshness: A★ 5KClaude Code
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get council-seat running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
council-seat. Agent Mart. https://agentmart.dev/templates/mock-server-mockserver-monorepo-council-seat (last updated 2026-10-24)