mickeyyaya/evolve-loop
Autonomous CI/CD for AI-written code: plans, builds, and adversarially audits changes to any repo — a different model reviews the builder — shipping only what passes deterministic gates. Failures become durable lessons; runs are checkpoint-
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit mickeyyaya/evolve-loop/.claude ./rig-evolve-loop # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit mickeyyaya/evolve-loop/skills/adversarial-testing .claude/skills/adversarial-testing $ npx degit mickeyyaya/evolve-loop/skills/architecture-review .claude/skills/architecture-review $ npx degit mickeyyaya/evolve-loop/skills/audit .claude/skills/audit $ npx degit mickeyyaya/evolve-loop/skills/code-review-simplify .claude/skills/code-review-simplify $ npx degit mickeyyaya/evolve-loop/skills/commit .claude/skills/commit $ npx degit mickeyyaya/evolve-loop/skills/diff-review .claude/skills/diff-review $ npx degit mickeyyaya/evolve-loop/skills/engineering-craft .claude/skills/engineering-craft $ npx degit mickeyyaya/evolve-loop/skills/evaluator .claude/skills/evaluator $ npx degit mickeyyaya/evolve-loop/skills/explain .claude/skills/explain $ npx degit mickeyyaya/evolve-loop/skills/fable .claude/skills/fable $ npx degit mickeyyaya/evolve-loop/skills/golang-test-review .claude/skills/golang-test-review $ npx degit mickeyyaya/evolve-loop/skills/inspirer .claude/skills/inspirer $ npx degit mickeyyaya/evolve-loop/skills/intent .claude/skills/intent $ npx degit mickeyyaya/evolve-loop/skills/loop .claude/skills/loop $ npx degit mickeyyaya/evolve-loop/skills/minimalism .claude/skills/minimalism $ npx degit mickeyyaya/evolve-loop/skills/phase-create .claude/skills/phase-create $ npx degit mickeyyaya/evolve-loop/skills/plan-review .claude/skills/plan-review $ npx degit mickeyyaya/evolve-loop/skills/publish .claude/skills/publish $ npx degit mickeyyaya/evolve-loop/skills/quality-index .claude/skills/quality-index $ npx degit mickeyyaya/evolve-loop/skills/refactor .claude/skills/refactor $ npx degit mickeyyaya/evolve-loop/skills/release .claude/skills/release $ npx degit mickeyyaya/evolve-loop/skills/retro .claude/skills/retro $ npx degit mickeyyaya/evolve-loop/skills/scout .claude/skills/scout $ npx degit mickeyyaya/evolve-loop/skills/security-review-scored .claude/skills/security-review-scored $ npx degit mickeyyaya/evolve-loop/skills/setup .claude/skills/setup $ npx degit mickeyyaya/evolve-loop/skills/ship .claude/skills/ship $ npx degit mickeyyaya/evolve-loop/skills/solution-audit .claude/skills/solution-audit $ npx degit mickeyyaya/evolve-loop/skills/solution-build .claude/skills/solution-build $ npx degit mickeyyaya/evolve-loop/skills/solution-scout .claude/skills/solution-scout $ npx degit mickeyyaya/evolve-loop/skills/tdd .claude/skills/tdd $ npx degit mickeyyaya/evolve-loop/skills/verify-release .claude/skills/verify-release
$ curl -fsSL --create-dirs -o .claude/agents/architecture-reviewer.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/.claude/agents/architecture-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/.claude/agents/code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/code-simplifier.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/.claude/agents/code-simplifier.md $ curl -fsSL --create-dirs -o .claude/agents/go-reviewer.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/.claude/agents/go-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/agent-templates.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/agent-templates.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-accessibility-audit.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-accessibility-audit.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-account-reconcile.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-account-reconcile.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-adversarial-review.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-adversarial-review.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-api-contract-design.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-api-contract-design.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-architecture-design.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-architecture-design.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-auditor-reference.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-auditor-reference.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-auditor.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-authz-gap-scan.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-authz-gap-scan.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-behavior-baseline.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-behavior-baseline.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-behavior-compare.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-behavior-compare.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-benchmark-gate.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-benchmark-gate.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-bug-reproduction.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-bug-reproduction.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-build-planner.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-build-planner.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-builder-reference.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-builder-reference.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-builder.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-builder.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-cache-strategy-scan.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-cache-strategy-scan.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-caching-strategy-design.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-caching-strategy-design.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-capacity-plan.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-capacity-plan.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-changelog-sync.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-changelog-sync.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-cicd-pipeline-audit.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-cicd-pipeline-audit.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-cleanup-sweep.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-cleanup-sweep.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-close-checklist.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-close-checklist.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-code-reviewer.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-compat-surface-check.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-compat-surface-check.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-container-hardening-scan.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-container-hardening-scan.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-context-condense.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-context-condense.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-contract-fuzz-probe.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-contract-fuzz-probe.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-coverage-gate.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-coverage-gate.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-data-integrity-check.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-data-integrity-check.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-data-model-design.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-data-model-design.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-debugger.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-debugger.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-deliverable-recovery.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-deliverable-recovery.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-dependency-audit.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-dependency-audit.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-dependency-map.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-dependency-map.md $ curl -fsSL --create-dirs -o .claude/agents/evolve-diagnose-reference.md https://raw.githubusercontent.com/mickeyyaya/evolve-loop/main/agents/evolve-diagnose-reference.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "B=\"$CLAUDE_PROJECT_DIR/go/bin/evolve\"; [ -x \"$B\" ] || B=\"$CLAUDE_PROJECT_DIR/go/evolve\"; \"$B\" guard ship --evolve-dir \"$CLAUDE_PROJECT_DIR/.evolve\""
},
{
"type": "command",
"command": "B=\"$CLAUDE_PROJECT_DIR/go/bin/evolve\"; [ -x \"$B\" ] || B=\"$CLAUDE_PROJECT_DIR/go/evolve\"; \"$B\" guard docdelete --evolve-dir \"$CLAUDE_PROJECT_DIR/.evolve\""
}
]
},
{
"matcher": "Agent|Task",
"hooks": [
{
"type": "command",
"command": "B=\"$CLAUDE_PROJECT_DIR/go/bin/evolve\"; [ -x \"$B\" ] || B=\"$CLAUDE_PROJECT_DIR/go/evolve\"; \"$B\" guard phase --evolve-dir \"$CLAUDE_PROJECT_DIR/.evolve\""
}
]
},
{
"matcher": "Edit|Write",
"hooks": [
{
"type": "command",
"command": "B=\"$CLAUDE_PROJECT_DIR/go/bin/evolve\"; [ -x \"$B\" ] || B=\"$CLAUDE_PROJECT_DIR/go/evolve\"; \"$B\" guard role --evolve-dir \"$CLAUDE_PROJECT_DIR/.evolve\""
}
]
}
]
}
} Skills (31)
adversarial-testingarchitecture-reviewauditcode-review-simplifycommitdiff-reviewengineering-craftevaluatorexplainfablegolang-test-reviewinspirerintentloopminimalismphase-createplan-reviewpublishquality-indexrefactorreleaseretroscoutsecurity-review-scoredsetupshipsolution-auditsolution-buildsolution-scouttddverify-release
Subagents (79)
| architecture-reviewer model: opus | Structural/architectural review of code changes — SOLID, dependency direction, design-pattern fit, SSOT/duplication, coupling — and whether the change was built test-first to the project's Clean Code |
| code-reviewer model: sonnet | Expert code reviewer for correctness, security, and maintainability of a change. Diff-scoped by design — reviews the changed hunks with bounded context, never reads whole packages or re-runs suites. U |
| code-simplifier model: sonnet | Simplifies recently modified code for clarity and consistency while preserving behavior exactly. Diff-scoped by design — works only inside the changed hunks, never reads packages or runs suites. Use a |
| go-reviewer model: sonnet | Expert Go code reviewer for idiomatic Go, error handling, concurrency, and test quality. Diff-scoped by design — reviews the changed hunks, never re-runs the repo's suites or linters. Use for all Go c |
| agent-templates | Base schemas and context blocks for all agents. |
| evolve-accessibility-audit model: tier-2 | Accessibility audit agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after build whenever scout.goal_type == "accessibility" to adversarially review changed UI/HTML/React |
| evolve-account-reconcile model: tier-2 | Account reconciliation agent for the Evolve Loop (Evaluate archetype). Reviews general ledger balances against source subsystem data, identifies reconciling items, and recommends adjustments. |
| evolve-adversarial-review model: tier-2 | Adversarial-review agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after build when the change touches source, to red-team the diff for exploitable weaknesses. Reasons l |
| evolve-api-contract-design model: tier-2 | Contract-first interface designer (Plan archetype) — authors the API surface contract before Builder touches implementation for feature cycles. |
| evolve-architecture-design model: tier-1 | Architecture-design agent for the Evolve Loop (Plan archetype). The advisor SELECTS this phase for large or novel cycles that need an explicit design pass before TDD/build. Reads scout's task + the co |
| evolve-auditor-reference | — |
| evolve-auditor model: tier-2 | Single-pass review agent for the Evolve Loop. Covers code quality, security, pipeline integrity, and eval gating. READ-ONLY — flags MEDIUM+ issues. |
| evolve-authz-gap-scan model: tier-1 | Authorization-gap adversary for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after build on security-relevant cycles (scout.security_relevant == true) to hunt for missing or in |
| evolve-behavior-baseline model: tier-2 | Behavior baseline agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase on refactor cycles before build (after TDD) to record a golden master of observable behavior and test o |
| evolve-behavior-compare model: tier-2 | Behavior comparison agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase on refactor cycles after build to compare post-build outputs against the baseline golden master and f |
| evolve-benchmark-gate model: tier-2 | Statistical benchmark comparison gate (Evaluate archetype). |
| evolve-bug-reproduction model: tier-2 | Bug reproduction agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase on bugfix cycles before build (after fault localization) to write a test or script that fails on the cur |
| evolve-build-planner model: tier-1 | Build-planning agent for the Evolve Loop. Moves Builder's design step into an independent phase (Opt C). Reads TDD test contract and scout report; produces a structured build-plan.md before Builder ex |
| evolve-builder-reference | — |
| evolve-builder model: tier-2 | Implementation agent for the Evolve Loop. Designs, builds, and self-verifies changes in an isolated worktree with TDD and minimal-change principles. |
| evolve-cache-strategy-scan model: tier-1 | Cache-coherence audit agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build on cycles whose scout.goal_type == "caching" to verify the implemented cache honors its |
| evolve-caching-strategy-design model: tier-1 | Caching-strategy designer for the Evolve Loop (Plan archetype). The advisor INSERTS this phase after Triage on caching cycles (scout.goal_type == "caching"), BEFORE any build, to commit the cache patt |
| evolve-capacity-plan model: tier-2 | Capacity planning agent for the Evolve Loop (Plan archetype). Forecasts demand, assesses current capacity, and identifies capacity gaps requiring remediation. |
| evolve-changelog-sync model: tier-1 | Changelog drift controller (Control archetype) — verifies CHANGELOG/release-notes match shipped commits after every cycle ship. |
| evolve-cicd-pipeline-audit model: tier-2 | CI/CD pipeline security gate for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build on infrastructure cycles (scout.goal_type == "infrastructure") to audit every changed |
| evolve-cleanup-sweep model: tier-2 | Dead-code and unused dependency detection (Evaluate archetype). |
| evolve-close-checklist model: tier-2 | Close checklist agent for the Evolve Loop (Control archetype). Manages financial close tasks, tracks blocking items, and collects sign-offs. |
| evolve-code-reviewer model: tier-1 | Independent code-review agent for the Evolve Loop (Evaluate archetype). The kernel pins this phase after Build on every code cycle whose build touched files. Plans a justified review of the build's di |
| evolve-compat-surface-check model: tier-1 | Backward-compatibility audit agent for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build whenever scout.goal_type == "api-design", to diff the realized exported surface |
| evolve-container-hardening-scan model: tier-2 | Container & manifest hardening gate for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build on infrastructure cycles to inspect the changed Dockerfiles and Kubernetes mani |
| evolve-context-condense model: tier-1 | Run-dir artifact condenser (Control archetype) — summarizes and prunes long per-cycle artifacts when cumulative size exceeds threshold to keep downstream phases within context budget. |
| evolve-contract-fuzz-probe model: tier-2 | Trust-boundary validation auditor for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after build on cycles where scout.goal_type == "api-design" to prove that changed input-handl |
| evolve-coverage-gate model: tier-3 | Coverage-regression gate for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build whenever the cycle's build.diff_loc is at least 50 changed lines, to verify the changed li |
| evolve-data-integrity-check model: tier-1 | Data-pipeline integrity auditor for the Evolve Loop (Evaluate archetype). The advisor INSERTS this phase after Build whenever the cycle's scout.goal_type == "data-pipeline", to statically audit the ch |
| evolve-data-model-design model: tier-1 | Data-model designer for the Evolve Loop (Plan archetype). The advisor INSERTS this phase after Triage on database cycles (scout.goal_type == "database"), BEFORE any build, to fix entities, keys, index |
| evolve-debugger model: opus | Ship-failure recovery diagnostician. Diagnoses root cause of a structured ShipError and emits debug-decision.json declaring RESHIP, RERUN_PHASE, or BLOCK. |
| evolve-deliverable-recovery model: tier-2 | The recovery agent of the logic-first ladder (ADR-0106). Dispatched when a phase's deliverable failed the contract gate on FORM only — a missing file, a missing section, a malformed JSON secondary — a |
| evolve-dependency-audit model: tier-2 | Dependency auditor for the Evolve Loop (Evaluate archetype). Reviews go.mod/go.sum changes for vulnerable, outdated, or incompatible dependencies. Emits dependency.severity_max; fails the cycle on >=C |
| evolve-dependency-map model: tier-2 | Dependency mapping and critical path analysis agent for the Evolve Loop (Evaluate archetype). Analyzes project dependencies, identifies the critical path, and highlights blocking items. |
| evolve-diagnose-reference | Layer-3 on-demand reference for diagnosing runtime failures in Evolve Loop. 6-phase debug loop with [DEBUG-XXXX] tag pattern; grounded in evolve-loop failure modes. |
Hooks (6)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Bash | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" guard ship --evolve-dir "$CLAUDE_PROJECT_DIR/.evolve" |
| PreToolUse | Agent|Task | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" guard phase --evolve-dir "$CLAUDE_PROJECT_DIR/.evolve" |
| PreToolUse | Edit|Write | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" guard role --evolve-dir "$CLAUDE_PROJECT_DIR/.evolve" |
| PreToolUse | Bash | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" guard docdelete --evolve-dir "$CLAUDE_PROJECT_DIR/.evolve" |
| PostToolUse | Edit|Write | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" postedit-validate |
| Stop | * | B="$CLAUDE_PROJECT_DIR/go/bin/evolve"; [ -x "$B" ] || B="$CLAUDE_PROJECT_DIR/go/evolve"; "$B" prune-ephemeral --quiet || true |
Slash commands (32)
/adversarial-testing/architecture-review/audit/build/code-review-simplify/commit/diff-review/engineering-craft/evaluator/explain/fable/golang-test-review/inspirer/intent/loop/minimalism/phase-create/plan-review/publish/quality-index/refactor/release/retro/scout/security-review-scored/setup/ship/solution-audit/solution-build/solution-scout/tdd/verify-release
Similar rigs
poshan0126/dotclaude
The standard .claude/ folder structure for everyday development.
Fort Knox 1.7k tok ·
kapadias/nonna
Your AI agent says done. Nonna makes it prove it: she runs your test suite before a coding agent can stop, and blocks pushes to main and secrets in files. Claude Code plugin, plus git hooks for Codex, Cursor, Copilot, Gemini and more
Fort Knox 3.9k tok ·
gufranco/claude-engineering-rules
Opinionated Claude Code engineering config: 11 always-on rules, 63 on-demand standards, 30 slash-command skills, 33 runtime hooks, 9 custom agents, 36 MCP servers, and a 774-item review checklist across 69 categories
YOLO Cowboy 156.1k tok ·
oprogramadorreal/optimus-claude
Primes your project for peak Claude Code performance
YOLO Cowboy 1.5k tok ·