cc10x

cc10x — The Loop Engine for Claude Code

Stop chasing better models. Engineer the loop.

Same model. Same Claude Code. Different outcome.
The agent forgets between turns. The harness remembers across runs.
cc10x writes every workflow to disk — intent, evidence, verdicts, failures — so resume, review, and verification read from the artifact, not from a context window that's already gone.

1 router  ·  9 specialist agents  ·  16 skills  ·  4 workflows

Fail-closed gates  ·  survives compaction  ·  dispatch-by-reference  ·  test honesty gates  ·  anti-anchored review

Current version: 12.5.0

---

Install

Step 1 — Add the marketplace:

/plugin marketplace add romiluz13/cc10x

Step 2 — Install the plugin:

/plugin install cc10x@cc10x

Then say "set up cc10x for me" in Claude Code and restart. Done.

---

Why cc10x

Ask Claude for something complex. It works for a while. Then it declares "Done!" — tests still red, refactor half-finished, and by message 40 it's contradicting itself because the context is gone.

cc10x fixes the loop, not the prompt. A better model running free loses to the same model, constrained and looped correctly. That's the whole bet.

The pain you know	How cc10x handles it
"Done!" on red tests	integration-verifier is independent of the builder. Phase-exit gates block advancement on partial evidence.
Silent failures nobody asked about	failure-hunter runs in parallel with review — greps for swallowed errors and empty catches.
Context falls apart after compaction	Workflow state on disk with stable UUIDs. Memory files the router auto-heals.
Planning is just a chat	Three planning modes chosen by intent, with a fresh anti-anchored review by a reviewer who never saw the planner's rationale.
12 slash commands to remember	One router. Every request hits cc10x-router first.
.claude/ prompt spam on every fanout	State lives at .cc10x/ — outside .claude/, so the harness's sensitive-file gate never fires.
Green tests that prove nothing	Test Honesty Gates grep for getByTestId('…-mock'), as any, .find() bypass, setTimeout() waits. A hit can't count as PASS on that test's strength alone.
Reviewer findings that sound right but aren't	Every finding at confidence ≥80 needs a verbatim file:line quote or it's auto-demoted. The verifier independently re-reads the line and drops hallucinated findings before they can gate.
Orchestrator context rotting with pasted history	Dispatch by reference, not by blob — the diff is written to disk, the prompt passes a path, never a body. (One real dispatch hit 42k chars, 99% pasted history. The scar became law.)

Everything below is the architecture. Keep reading if you want to know how before you install.

---

Runtime Model

1. Router owns orchestration

cc10x-router is the only orchestration authority.

The router now uses a kernel + mandatory reference shape:

• universal orchestration law stays inline in cc10x-router/SKILL.md
• workflow-specific playbooks and appendix-heavy artifact/remediation law live in cc10x-router/references/*.md
• the kernel explicitly tells Claude which reference must be read before BUILD / DEBUG / REVIEW / PLAN branch logic continues

That keeps orchestration salient without turning the router into a context dump.

It decides:

• which workflow to run
• which subagent to invoke next
• when to pause for clarification or scope decisions
• when remediation is required
• when a workflow is allowed to advance
• when memory and workflow artifacts are finalized

Agents do not own workflow state. They return structured results. The router interprets them.

2. Agents are narrow specialists

The shipped subagents are intentionally specialized:

• planner
• plan-gap-reviewer
• component-builder
• bug-investigator
• code-reviewer
• failure-hunter
• integration-verifier
• researcher
• doc-syncer

Each agent is optimized for one role. This keeps prompts sharper and makes workflow behavior easier to reason about.

3. Skills are reusable local instructions

Skills are the reusable instruction layer that agents and the router depend on.

They provide:

• planning patterns
• TDD rules
• debugging patterns
• review rules
• research synthesis
• memory handling
• verification-before-completion discipline

4. Workflow artifacts are the durable truth

cc10x writes proof-of-work workflow state under: