rustcircle/c101
Multi-agent all gathered on a table to discuss lancelot with agentic workflows
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
GUARDRAILS
3/5
Protects secrets · Pre-tool screening hook · No YOLO mode · details
Quality 18/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit rustcircle/c101/.claude ./rig-c101 # 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 rustcircle/c101/.claude/skills/adversarial-verification .claude/skills/adversarial-verification $ npx degit rustcircle/c101/.claude/skills/algorithm-design-paradigms .claude/skills/algorithm-design-paradigms $ npx degit rustcircle/c101/.claude/skills/amortized-recurrence-analysis .claude/skills/amortized-recurrence-analysis $ npx degit rustcircle/c101/.claude/skills/analogical-transfer .claude/skills/analogical-transfer $ npx degit rustcircle/c101/.claude/skills/api-contract-design .claude/skills/api-contract-design $ npx degit rustcircle/c101/.claude/skills/automata-regular-languages .claude/skills/automata-regular-languages $ npx degit rustcircle/c101/.claude/skills/bayesian-aggregation .claude/skills/bayesian-aggregation $ npx degit rustcircle/c101/.claude/skills/brand-guidelines-system .claude/skills/brand-guidelines-system $ npx degit rustcircle/c101/.claude/skills/browser-automation .claude/skills/browser-automation $ npx degit rustcircle/c101/.claude/skills/bvh-build-traversal .claude/skills/bvh-build-traversal $ npx degit rustcircle/c101/.claude/skills/caching-and-system-tradeoffs .claude/skills/caching-and-system-tradeoffs $ npx degit rustcircle/c101/.claude/skills/cargo-workspaces-features .claude/skills/cargo-workspaces-features $ npx degit rustcircle/c101/.claude/skills/chart-design-encoding .claude/skills/chart-design-encoding $ npx degit rustcircle/c101/.claude/skills/ci-pipelines .claude/skills/ci-pipelines $ npx degit rustcircle/c101/.claude/skills/cirq-workflows .claude/skills/cirq-workflows $ npx degit rustcircle/c101/.claude/skills/cmake-presets .claude/skills/cmake-presets $ npx degit rustcircle/c101/.claude/skills/code-quality-metrics .claude/skills/code-quality-metrics $ npx degit rustcircle/c101/.claude/skills/code-review-checklist .claude/skills/code-review-checklist $ npx degit rustcircle/c101/.claude/skills/codemod-automation .claude/skills/codemod-automation $ npx degit rustcircle/c101/.claude/skills/coding-general .claude/skills/coding-general $ npx degit rustcircle/c101/.claude/skills/color-pipeline-tonemapping .claude/skills/color-pipeline-tonemapping $ npx degit rustcircle/c101/.claude/skills/combinatorics-generating-functions .claude/skills/combinatorics-generating-functions $ npx degit rustcircle/c101/.claude/skills/compiler-construction .claude/skills/compiler-construction $ npx degit rustcircle/c101/.claude/skills/computer-graphics .claude/skills/computer-graphics $ npx degit rustcircle/c101/.claude/skills/computer-use .claude/skills/computer-use $ npx degit rustcircle/c101/.claude/skills/concurrent-data-structures .claude/skills/concurrent-data-structures $ npx degit rustcircle/c101/.claude/skills/concurrent-fanout .claude/skills/concurrent-fanout $ npx degit rustcircle/c101/.claude/skills/container-images .claude/skills/container-images $ npx degit rustcircle/c101/.claude/skills/copyediting .claude/skills/copyediting $ npx degit rustcircle/c101/.claude/skills/cpp-concurrency-atomics .claude/skills/cpp-concurrency-atomics $ npx degit rustcircle/c101/.claude/skills/cpp-craft .claude/skills/cpp-craft $ npx degit rustcircle/c101/.claude/skills/cpp-memory-raii .claude/skills/cpp-memory-raii $ npx degit rustcircle/c101/.claude/skills/cpp-simd-performance .claude/skills/cpp-simd-performance $ npx degit rustcircle/c101/.claude/skills/cpp-templates-concepts .claude/skills/cpp-templates-concepts $ npx degit rustcircle/c101/.claude/skills/cpp-testing-fuzzing .claude/skills/cpp-testing-fuzzing $ npx degit rustcircle/c101/.claude/skills/cpp-ub-sanitizers .claude/skills/cpp-ub-sanitizers $ npx degit rustcircle/c101/.claude/skills/cs-foundations .claude/skills/cs-foundations $ npx degit rustcircle/c101/.claude/skills/cudaq-gpu-kernels .claude/skills/cudaq-gpu-kernels $ npx degit rustcircle/c101/.claude/skills/data-structure-selection .claude/skills/data-structure-selection $ npx degit rustcircle/c101/.claude/skills/database-transactions-isolation .claude/skills/database-transactions-isolation
$ curl -fsSL --create-dirs -o .claude/agents/computer-graphics-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/computer-graphics-expert.md $ curl -fsSL --create-dirs -o .claude/agents/computer-science-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/computer-science-expert.md $ curl -fsSL --create-dirs -o .claude/agents/cpp-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/cpp-expert.md $ curl -fsSL --create-dirs -o .claude/agents/go-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/go-expert.md $ curl -fsSL --create-dirs -o .claude/agents/graphic-design-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/graphic-design-expert.md $ curl -fsSL --create-dirs -o .claude/agents/haskell-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/haskell-expert.md $ curl -fsSL --create-dirs -o .claude/agents/java-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/java-expert.md $ curl -fsSL --create-dirs -o .claude/agents/julia-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/julia-expert.md $ curl -fsSL --create-dirs -o .claude/agents/kotlin-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/kotlin-expert.md $ curl -fsSL --create-dirs -o .claude/agents/mathematics-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/mathematics-expert.md $ curl -fsSL --create-dirs -o .claude/agents/physics-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/physics-expert.md $ curl -fsSL --create-dirs -o .claude/agents/python-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/python-expert.md $ curl -fsSL --create-dirs -o .claude/agents/quantum-computing-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/quantum-computing-expert.md $ curl -fsSL --create-dirs -o .claude/agents/rust-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/rust-expert.md $ curl -fsSL --create-dirs -o .claude/agents/typescript-expert.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/experts/typescript-expert.md $ curl -fsSL --create-dirs -o .claude/agents/image-editor.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/image/image-editor.md $ curl -fsSL --create-dirs -o .claude/agents/image-generator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/image/image-generator.md $ curl -fsSL --create-dirs -o .claude/agents/vector-artist.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/image/vector-artist.md $ curl -fsSL --create-dirs -o .claude/agents/visual-reviewer.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/image/visual-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/agent-architect.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/meta/agent-architect.md $ curl -fsSL --create-dirs -o .claude/agents/stack-tuner.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/meta/stack-tuner.md $ curl -fsSL --create-dirs -o .claude/agents/concurrent-dispatcher.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/concurrent-dispatcher.md $ curl -fsSL --create-dirs -o .claude/agents/delegator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/delegator.md $ curl -fsSL --create-dirs -o .claude/agents/ensemble-assembler.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/ensemble-assembler.md $ curl -fsSL --create-dirs -o .claude/agents/mediator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/mediator.md $ curl -fsSL --create-dirs -o .claude/agents/mesh-router.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/mesh-router.md $ curl -fsSL --create-dirs -o .claude/agents/orchestrator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/orchestrator.md $ curl -fsSL --create-dirs -o .claude/agents/swarm-coordinator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/orchestration/swarm-coordinator.md $ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/planning/architect.md $ curl -fsSL --create-dirs -o .claude/agents/plan-critic.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/planning/plan-critic.md $ curl -fsSL --create-dirs -o .claude/agents/planner.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/planning/planner.md $ curl -fsSL --create-dirs -o .claude/agents/adversarial-verifier.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/adversarial-verifier.md $ curl -fsSL --create-dirs -o .claude/agents/code-optimizer.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/code-optimizer.md $ curl -fsSL --create-dirs -o .claude/agents/code-quality-analyst.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/code-quality-analyst.md $ curl -fsSL --create-dirs -o .claude/agents/debugger.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/debugger.md $ curl -fsSL --create-dirs -o .claude/agents/quality-control.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/quality-control.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/security-auditor.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/security-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/tester.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/tester.md $ curl -fsSL --create-dirs -o .claude/agents/validator.md https://raw.githubusercontent.com/rustcircle/c101/main/.claude/agents/quality/validator.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Read(~/.claude/stack.env)",
"Bash(cat ~/.claude/stack.env*)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Agent|Task",
"hooks": [
{
"type": "command",
"command": "\"$CLAUDE_PROJECT_DIR\"/bin/stack hook pre-agent"
}
]
},
{
"matcher": "Bash|Write|Edit|MultiEdit|NotebookEdit|mcp__playwright__browser_type|mcp__playwright__browser_fill_form|mcp__playwright_",
"hooks": [
{
"type": "command",
"command": "\"$CLAUDE_PROJECT_DIR\"/bin/stack hook pre-tool"
}
]
}
]
}
} Skills (150)
adversarial-verificationalgorithm-design-paradigmsamortized-recurrence-analysisanalogical-transferapi-contract-designautomata-regular-languagesbayesian-aggregationbrand-guidelines-systembrowser-automationbvh-build-traversalcaching-and-system-tradeoffscargo-workspaces-featureschart-design-encodingci-pipelinescirq-workflowscmake-presetscode-quality-metricscode-review-checklistcodemod-automationcoding-generalcolor-pipeline-tonemappingcombinatorics-generating-functionscompiler-constructioncomputer-graphicscomputer-useconcurrent-data-structuresconcurrent-fanoutcontainer-imagescopyeditingcpp-concurrency-atomicscpp-craftcpp-memory-raiicpp-simd-performancecpp-templates-conceptscpp-testing-fuzzingcpp-ub-sanitizerscs-foundationscudaq-gpu-kernelsdata-structure-selectiondatabase-transactions-isolationdebugging-methoddeep-researchdelegation-protocoldelphi-methoddenoising-taa-upscalingdependency-upgradesdesign-tokens-dtcgdevelopmental-editingdimensional-analysis-unitsdistributed-consensus-consistencydocs-and-adr-writingdoubt-escalationdynamic-programming-patternseffort-and-model-selectionem-field-solvers-fdtd-femensemble-assemblyexpert-synthesisfastapi-pydantic-servicesfiction-craftfirst-principlesfluid-dynamics-cfdformal-verification-tla-dafnygit-advanced-workflowsgo-concurrency-contextgo-craftgo-error-handlinggo-http-servicesgo-interfaces-genericsgo-modules-toolinggo-profiling-pprofgo-testing-fuzzinggpu-compute-patternsgpu-debugging-profilinggraph-algorithmsgraph-theory-probabilistic-methodgraphic-designhamiltonian-simulation-methodshaskell-cabal-buildshaskell-concurrency-stmhaskell-crafthaskell-effectshaskell-performance-space-leakshaskell-testing-propertieshaskell-type-levelicon-system-designimage-brief-writingimage-editingimage-generationimportance-sampling-mis-restirinequality-provingjava-build-maven-gradlejava-craftjava-jvm-tuning-jfrjava-modern-languagejava-quarkusjava-spring-bootjava-testing-junitjava-virtual-threadsjudge-paneljulia-craftjulia-gpujulia-ml-flux-luxjulia-optimization-jumpjulia-packages-environmentsjulia-performance-allocationsjulia-sciml-diffeqjulia-type-stabilitykotlin-composekotlin-coroutines-flowkotlin-craftkotlin-gradle-buildskotlin-java-interopkotlin-multiplatformkotlin-testinglagrangian-hamiltonian-mechanicslayout-grids-posters-editoriallean4-mathlib-formalizationliterature-reviewlogo-design-processmanuscript-projectmathematics-toolkitmatrix-decompositions-conditioningmediator-votingmesh-blackboardmesh-processingmetal-apple-gpumolecular-dynamics-setupnode-runtime-patternsnp-hardness-reductionsnumber-theory-modularnumerical-error-analysisobservability-instrumentationode-pde-analytic-methodsoklch-palette-rampsoptics-wave-propagationoptimization-convex-nonconvexorchestration-patternsoutput-contractpath-tracer-implementationpbr-microfacet-brdfpde-discretization-stabilitypennylane-differentiableperformance-optimizationphoto-retouch-compositephysics-ode-integratorsphysics-toolkitplanning-dagpolishprint-prepress-pdfxprobability-distributions-limits
Subagents (63)
| computer-graphics-expert model: claude-opus-5-5 | Rendering, GPU programming and geometry expert that designs, implements, debugs and validates path tracers, real-time renderers, shaders (GLSL/HLSL/WGSL/MSL/Slang), BRDFs and sampling, denoisers, mesh |
| computer-science-expert model: claude-opus-5-5 | Theoretical and systems computer science expert for algorithms and data structures, complexity and reductions, automata and parsing, type systems and PL semantics, compilers, distributed consensus, da |
| cpp-expert model: claude-opus-5-5 | Modern C++ (C++20/23, C++26 where the toolchain supports it) specialist that designs, implements, reviews and hardens code with RAII and value semantics, concepts and constexpr, ranges, coroutines, at |
| go-expert model: claude-sonnet-5-5 | Go specialist that builds, reviews, debugs and tunes idiomatic Go services and libraries (small interfaces, wrapped errors, generics and iterators, goroutines with context and errgroup, net/http and g |
| graphic-design-expert model: claude-opus-5-5 | Art director and graphic designer that defines composition, typography, color systems, layout grids, iconography and brand rules, writes precise image briefs, drives the image MCP server's configured |
| haskell-expert model: claude-opus-5-5 | Haskell and typed functional programming specialist that designs, implements, reviews and tunes code with type classes and laws, GADTs, type families and DataKinds, controlled laziness and strictness, |
| java-expert model: claude-sonnet-5-5 | Modern Java (21 to 25) specialist that designs, implements, reviews and tunes JVM code with records, sealed types and pattern matching, virtual threads, streams and collections, Spring Boot or Quarkus |
| julia-expert model: claude-sonnet-5-5 | Julia specialist for scientific and numerical computing that designs, implements, reviews and speeds up code using multiple dispatch, type-stable generic functions, allocation-free broadcasting and lo |
| kotlin-expert model: claude-sonnet-5-5 | Kotlin specialist (JVM, Multiplatform, Android) that designs, implements, reviews and debugs code with null safety, data, value and sealed types, coroutines and Flow under structured concurrency, type |
| mathematics-expert model: claude-opus-5-5 | Mathematics expert that solves problems and writes rigorous proofs across algebra, analysis, probability and statistics, combinatorics, linear algebra, optimisation, numerical analysis and geometry, s |
| physics-expert model: claude-opus-5-5 | Physics expert for classical and quantum mechanics, electromagnetism, statistical mechanics, special and general relativity, fluids, condensed matter and computational physics, who models a situation, |
| python-expert model: claude-sonnet-5-5 | Python specialist that builds, reviews, debugs and tunes typed, idiomatic Python (strict mypy or pyright, asyncio with TaskGroup, uv-managed pyproject and lockfiles, pytest, FastAPI and pydantic v2) a |
| quantum-computing-expert model: claude-opus-5-5 | Quantum computing expert for circuits and algorithms (QFT, phase estimation, Grover and amplitude amplification, VQE, QAOA, Hamiltonian simulation), error correction and decoding (surface and qLDPC co |
| rust-expert model: claude-opus-5-5 | Rust specialist that writes, reviews, debugs and optimizes idiomatic, safe, fast Rust (ownership and lifetimes, traits and generics, tokio async, unsafe and FFI audits, no_std, cargo workspaces) and p |
| typescript-expert model: claude-sonnet-5-5 | TypeScript and Node specialist that builds, reviews, debugs and migrates strictly typed code (unions, satisfies, ESM/CJS, package exports, AbortSignal async, React hooks, Node services) and proves it |
| image-editor model: claude-sonnet-5-5 | Edits existing images through the image MCP server's configured operations (EDIT instruction edits and masked inpainting or outpainting, SEGMENT prompt-based masks, BG_REMOVAL cutouts to clean RGBA, U |
| image-generator model: claude-sonnet-5-5 | Generates new raster images through the image MCP server's configured operations (GEN text-to-image, REF multi-reference generation that keeps a character, product or style consistent, TRANSPARENT nat |
| vector-artist model: claude-sonnet-5-5 | Produces clean, safe SVG through the image MCP server (VECTOR native generation of icons, logos and flat illustrations, TRACE raster-to-vector tracing, SVG manipulation by editing the SVG text itself |
| visual-reviewer model: claude-opus-5-5 | Judges raster images, edits, cutouts, masks, upscales, SVG vectors and layered SVG/PDF deliverables against their brief with a weighted 0-10 rubric (brief fidelity, composition and hierarchy, typograp |
| agent-architect model: claude-opus-5-5 | Designs and writes new agents, skills and saved workflows for this stack to its conventions (frontmatter keys and order, single-line routing description, body sections, delegation and depth rules, out |
| stack-tuner model: claude-sonnet-5-5 | Operates the stack's Bayesian self-tuning loop - runs bin/stack learn ingest, update, report and apply, reads the per-agent posteriors (censored log-normal turns, effort bands, failure rate, token win |
| concurrent-dispatcher model: claude-haiku-5-5 | Pure fan-out/fan-in executor - given a list of independent work items, an agent type and a prompt template, it launches them concurrently in batches of up to 20 Agent calls per message, collects struc |
| delegator model: claude-opus-5-5 | Main-thread router (depth 0) that classifies every request, picks the cheapest adequate route, writes a short routing plan, dispatches to orchestrators, experts, researchers or saved workflows, and ke |
| ensemble-assembler model: claude-opus-5-5 | Assembles an ad-hoc team of 2-7 expert agents for a problem that spans domains - decomposes it into disciplines, picks members with explicit roles (lead, specialist, challenger, integrator), writes ea |
| mediator model: claude-opus-5-5 | Mediates collective decisions - has N diverse agents (different agent types and tiers, perspectives) solve the same problem independently, anonymises their answers, runs blind cross-critique, applies |
| mesh-router model: claude-sonnet-5-5 | Runs a sub-agent mesh over a shared blackboard directory (.stack/blackboard/<task-id>/) where specialists post claims, evidence, questions, answers and objections as files; each round it reads the boa |
| orchestrator model: claude-opus-5-5 | Multi-phase project executor that turns a goal into a task DAG (plan, implement, verify), dispatches each node to the best specialist, runs independent nodes concurrently in waves, threads artifacts b |
| swarm-coordinator model: claude-opus-5-5 | Runs swarms over large jobs by partitioning many files, sources, hypotheses or parameter points into shards, launching many workers concurrently (same or mixed agent types), retrying failed shards, de |
| architect model: claude-opus-5-5 | Advanced planner for system and software architecture - turns quality-attribute requirements into 2-4 structurally different candidate designs (MVP-first, risk-first, performance-first, evolvability-f |
| plan-critic model: claude-opus-5-5 | Red-teams plans, task DAGs and architecture decisions - checks goal fit, traceability, falsifiability of acceptance criteria, hidden dependencies, write-set conflicts, missing steps (migration, rollba |
| planner model: claude-opus-5-5 | Decomposes a goal into an executable plan file - a task DAG with ids, dependencies, owner agents, write-sets, falsifiable acceptance criteria, verification steps, risks, parallel waves and the critica |
| adversarial-verifier model: claude-opus-5-5 | Skeptic that tries to refute one claim - a review finding, bug report, proof step, performance claim or research assertion - by restating it as a falsifiable proposition, hunting for guards and counte |
| code-optimizer model: claude-opus-5-5 | Performance engineer that profiles before touching code (benchmarks with statistics, CPU and off-CPU flamegraphs, allocation and heap profiles), forms hypotheses sized by Amdahl's law, applies algorit |
| code-quality-analyst model: claude-sonnet-5-5 | Measures code quality with real tools where available (linters, cyclomatic and cognitive complexity, duplication, dependency graphs and cycles, type coverage, git churn hotspots) and by reading the ho |
| debugger model: claude-opus-5-5 | Root-cause analyst for bugs, crashes, flaky tests and regressions. Reproduces the failure deterministically, minimises the input, bisects commits, dependencies or config, instruments with debuggers, s |
| quality-control model: claude-opus-5-5 | Final quality gate before a result reaches the user. Builds a requirement-to-evidence traceability matrix against the original request, re-runs cheap verifications, flags unverified load-bearing claim |
| reviewer model: claude-opus-5-5 | Advanced code reviewer for diffs, branches and files - reads every hunk in context with its callers, runs the project's own compiler, linter and tests on touched packages, and reports findings across |
| security-auditor model: claude-opus-5-5 | Defensive security reviewer that threat-models the target (STRIDE per trust boundary), traces untrusted input from sources to sinks, and audits input validation, authentication and authorisation, inje |
| tester model: claude-sonnet-5-5 | Designs, writes and runs tests - unit, property-based, fuzz, metamorphic, differential, integration, regression, Playwright E2E and desktop GUI E2E in a disposable network-less sandbox - writing a min |
| validator model: claude-sonnet-5-5 | Mechanically checks a deliverable against its spec - turns acceptance criteria into a criteria matrix, runs the build, lint, type-check and test commands, validates schemas, invariants and numeric tol |
Hooks (9)
| event | matcher | runs |
|---|---|---|
| SessionStart | * | "$CLAUDE_PROJECT_DIR"/bin/stack hook session-start |
| PreToolUse | Agent|Task | "$CLAUDE_PROJECT_DIR"/bin/stack hook pre-agent |
| PreToolUse | Bash|Write|Edit|MultiEdit|NotebookEdit|mcp__playwright__browser_type|mcp__playwright__browser_fill_form|mcp__playwright_ | "$CLAUDE_PROJECT_DIR"/bin/stack hook pre-tool |
| SubagentStop | * | "$CLAUDE_PROJECT_DIR"/bin/stack hook subagent-stop |
| SessionEnd | * | "$CLAUDE_PROJECT_DIR"/bin/stack hook session-end |
| UserPromptSubmit | * | "$CLAUDE_PROJECT_DIR"/bin/stack hook prompt-submit |
| PreModelSwitch | * | "$CLAUDE_PROJECT_DIR"/bin/stack hook model-switch |
| PostToolUse | Agent|Task|Workflow | "$CLAUDE_PROJECT_DIR"/bin/stack hook post-agent |
| PostToolUse | WebFetch|WebSearch|mcp__fetch__.*|mcp__jina__.*|mcp__exa__.*|mcp__spider__.*|mcp__playwright__.*|mcp__firecrawl__.*|mcp_ | "$CLAUDE_PROJECT_DIR"/bin/stack hook post-web |
Permissions
deny (2)
Read(~/.claude/stack.env)
Bash(cat ~/.claude/stack.env*)
ask (0)
—
allow (19)
Bash(bin/stack *)
Bash(./bin/stack *)
mcp__exa
mcp__jina
mcp__spider
mcp__image
mcp__context7
mcp__desktop-sandbox
Workflow(swarm-run)
Workflow(fanout)
Workflow(mediate)
Workflow(expert-panel)
Workflow(research-sweep)
Workflow(review-verify)
Workflow(plan-tournament)
Workflow(visual-iterate)
Workflow(manuscript-review)
mcp__fetch
mcp__neural-memory
Similar rigs
claude-dev-suite/claude-dev-suite
Point it at a repo and it configures your AI coding assistant for that stack: agents, framework skills, MCP servers, path-scoped rules. One install, seven assistants - Claude Code, Copilot, Cursor, Gemini CLI, Codex CLI, Cline, Kimi Code. M
Orchestrator 9.3k tok · · Q3
qyndex/claude-init
Production-grade Claude Code harness — a spec-driven, TDD-first, verification-gated, security-aware .claude/ scaffold you drop into any repo with one command.
Fort Knox 98.6k tok · · Q90
AgentWorkforce/relay
Infrastructure for coding agents
Orchestrator 8.0k tok · · Q18
fjerbi/claude-code-arsenal
A tested Claude Code operating system for efficient, reliable, and focused software engineering. Includes global instructions, workflows, rules, and tools designed to minimize wasted tokens, prevent hallucinations and unnecessary loops, enf
YOLO Cowboy 4.0k tok · · Q51