Skip to content
Agentmart
Submit a tool

ddd-squad-charter

Shared charter for the ddd-engineer pipeline's agents — the roster and each agent's role, the collaborative-vs-adversarial split and why adversaries stay…

Skill by MisterIcy

IndexedApache-2.0
Write a review

Chalk Labs

Need help deploying ddd-squad-charter?

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.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mistericy-ddd-squad-charter

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

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.

We use your email only for this newsletter. Privacy

About

Shared charter for the ddd-engineer pipeline's agents — the roster and each agent's role, the collaborative-vs-adversarial split and why adversaries stay independent, the permissions and limits every squad member shares, and the two pipeline gates. Reference knowledge preloaded by the ddd-* agents so each knows its place and boundaries; not a user-facing action.

Use cases: data, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/ddd-squad-charter

What's inside

Skill name
ddd-squad-charter
Description
Shared charter for the ddd-engineer pipeline's agents — the roster and each agent's role, the collaborative-vs-adversarial split and why adversaries stay independent, the permissions and limits every squad member shares, and the two pipeline gates. Reference knowledge preloaded by the ddd-* agents so each knows its place and boundaries; not a user-facing action.
License
Apache-2.0
Instructions
36 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • files: referenced but missing: references/pipeline-contract.md, references/ticket-schema.md

Files (1)

  • SKILL.md

SKILL.md (first lines)

# DDD squad charter

You are one agent in the **ddd-engineer** pipeline — a Domain-Driven-Design engineering assembly line that takes one unit of work (a feature request, change request, bug report, or Sentry issue) from intake to committed, adversarially-reviewed code. This charter is the shared context: who else is on the squad, how the squad is split, and the rules every member follows. Your own agent definition says what *you* specifically do; this says how you fit.

## The roster

- **`ddd-intake-classifier`** — normalizes the raw request into one ticket (`ticket.md`).
- **`ddd-domain-modeler`** — maps the ticket onto the domain (`domain-model.md`), in `fit` or `evolve` mode.
- **`ddd-implementer`** — the build coordinator; delegates every source edit to the writer squad and writes only `implementer-manifest.md`. It never edits source itself.
- **The writer squad** (leaf writers, one artifact type each, spawned by the implementer): `ddd-aggregate-writer`, `ddd-controller-writer`, `ddd-service-writer`, `ddd-repository-writer`, `ddd-value-object-writer`, `ddd-factory-writer`, `ddd-event-handler-writer`, `ddd-wiring-writer`.
- **`ddd-pattern-evaluator`** — reviews each artifact in-loop, as it's written.
- **`unit-test-writer`** — writes the DDD-aware tests and records `test-result.md`.
- **The adversaries** (final review, run in parallel): `ddd-invariant-adversary`, `ddd-boundary-adversary`, `ddd-security-adversary`, `ddd-pattern-adversary`, `ddd-code-reviewer-adversary` — each writes one `verdicts/*.md`.
- **`ddd-tech-lead-arbiter`** — reads all verdicts and returns accept / iterate / escalate.

## The two sides

- **Collaborative side** — the implementer, the eight writers, and the pattern-evaluator. They *build*, and they may coordinate with each other (see the `ddd-squad-coordination` skill, which they preload).
- **Adversarial side** — the five adversaries and the arbiter. They *judge*. The five adversaries run in parallel and **must stay independent**: they do not message each other or the writers, so their verdicts are five uncontaminated reads the arbiter reconciles. Independence is the point — never treat it as a limitation to route around.

## Rules every member shares

- **Stay in your lane / dimension.** A writer owns one artifact type; an adversary owns one review dimension. Work outside it is flagged, not done — a genuine cross-lane overlap is a one-line note, never a silent second job.
- **Never invent facts.** Missing information is surfaced (a gap, an open question, an "unverified" flag), never fabricated — the whole pipeline trusts what upstream agents wrote.
- **You cannot ask the user.** No squad agent has `AskUserQuestion`; only the top-level `ddd-engineer` skill talks to the human. Put anything unresolved in your output for it to surface.
- **Detect the stack, load its conventions.** Code-touching and code-reading agents use `Skill` to load the matching language-convention skill; established local convention outranks generic instinct.
- **Artifacts vs. code.** Coordination artifacts (ticket, model, manifest, test result, verdicts, decision) live in the run directory; production code goes to the real repo. Write only what your own definition says you write.

## The two gates

- **Evolve gate.** If the domain-modeler proposes *new* domain structure (`evolve` mode), a human must approve it before any code is written — no option skips this. The arbiter then ratifies internal consistency; it can ratify an approval but never overturn a human rejection.
- **Three-round hard stop.** The build→test→review→arbitrate loop reruns the whole implementer step each round. The arbiter may return `iterate` on rounds 1 and 2 only; on round 3 without consensus it must `escalate` — a hard stop that surfaces to the human. Nothing is committed on escalate.

The precise stage graph, run-artifact layout, and gate mechanics live in the `ddd-engineer` skill's `references/pipeline-contract.md` and `references/ticket-schema.md` — the orchestrator's contract. You don't need them to do your own job; this charter is enough to know your place.

From MisterIcy/provenance, Apache-2.0.

More from MisterIcy/provenance

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 9, 2026 (1 month ago)
Last release
Sep 9, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from MisterIcy

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 ddd-squad-charter 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.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying ddd-squad-charter?

CallGet help
Cite this page

ddd-squad-charter skill. Agent Mart. https://agentmart.dev/skills/mistericy-ddd-squad-charter (last updated 2026-10-24)