33
Published Tools
3
Total Stars
4,849
Weekly Downloads
1
GitHub Followers
14
Public Repos
100/100
Avg Security
Published Tools
12 MCP Servers20 Skills1 Agentacross 9 categoriesMax Meier
Track H / §59 — emits an Anthropic-compatible Claude Code plugin directory from any CrewHaus IR variant. Source: claude-plugins-official (Anthropic).
...morecrewhaus
MCP server for CrewHaus startup validation services. Lets AI agents discover, price, and submit validation requests.
AI startup idea validation by a specialized 9-agent crew.
crewhaus
MCP server for CrewHaus Agent Certifications. Thin client over certify.crewhaus.ai API.
crewhaus
Backend-agnostic spec-template registry: git/huggingface/npm/local backends + TTL cache + sigstore-style signature verification (Section 40)
...moreMax Meier
Anthropic provider adapter + shared ProviderAdapter interface (Section 17)
Max Meier
OpenAI provider adapter (Section 17). Chat Completions streaming + function calls
Max Meier
MCP client manager: stdio + SSE transports, exp-backoff reconnect, server registry, built on @modelcontextprotocol/sdk
Max Meier
Wrap remote MCP tools as RegisteredTool entries via @crewhaus/tool-builder
Max Meier
Pillar-2 model-driven MutationProvider — asks Claude to rewrite the prompt given dev-set failures. Closes the v0 prompt-optimizer's L91 'rule-based-only' gap.
...moreMax Meier
Vector index — in-memory flat L2 default, file-backed lance + HTTP backends (qdrant/pinecone/weaviate)
Max Meier
Pillar-3 model-backed JustificationJudge — asks Claude whether a tool call's justification is genuinely consistent with the compile-time session goal. Ships the production judge the intent gate's footnote promised; fails closed on model error.
...moreMax Meier
Project a compiled bundle's turn function as an MCP server (stdio + SSE) — a chat/invoke tool plus optional per-sub-agent tools delegating to an injected invoke fn, built on @modelcontextprotocol/sdk
...morestudiomaxops
ReadImage tool — validates magic bytes and returns Anthropic base64 image content blocks
studiomaxops
FindElement(description) tool — vision-grounded bounding-box lookup (Section 25 BROW)
studiomaxops
CrewHaus — the meta-harness compiler for AI agents. Compile one crewhaus.yaml spec into a CLI agent, channel bot, RAG pipeline, multi-agent crew, eval harness, voice or browser agent, and more.
...morestudiomaxops
Resolve agent.model strings to a lazy-loaded ProviderAdapter (Section 17)
studiomaxops
AWS Bedrock provider adapter (Section 17). Anthropic over native InvokeModel; all other families (Llama / Mistral / Nova / DeepSeek / Cohere / AI21 / Qwen / gpt-oss / Titan / Writer) over the Converse API
...morestudiomaxops
Embedding provider — openai/voyage/cohere/local/mock backends with prefix-grammar selector
studiomaxops
Citation-tracker-backed crawler — HTTP fetch + file:// reads with URL-dedup, rate limiting (Section 23 RES)
studiomaxops
Eval dataset loaders (JSONL, CSV, YAML, HTTP) with a lazy iterator API
studiomaxops
Deterministic eval-dataset operations: version, inspect, validate and grow a dataset from local files and recorded sessions
...morestudiomaxops
Split-aware dataset versioning: train/dev/test with hash-stable sample identifiers and split-leak prevention
studiomaxops
Eval dataset operations: reference resolution and splits, lint, PII/secret audit, mining and synthesis from session logs, and grader suggestion from failures
...morestudiomaxops
Pillar-3 chokepoint — classify content crossing every trust boundary (MCP / sub-agent / channel / federation / skill / compaction / tool) with origin metadata and a content-hash LRU cache
...morestudiomaxops
Cross-cutting safety classifier for tool outputs (regex + structural heuristics + optional LLM tier)
studiomaxops
Pillar-3 sink-side chokepoint — classify content leaving via external sinks (fetch / web / mcp / channel / federation / evm-tx) against the data-lineage carried in run-context
...morestudiomaxops
Composable PII redaction: regex + classifier + policy-driven detectors; replace and hash modes (Section 39)
studiomaxops
Safety-classifier graders: toxicity, bias, pii_leak. Pluggable via @crewhaus/grader-registry (Section 38)
studiomaxops
Deterministic validation tools: JSON Schema validation and inference, JSON Patch, deep diffing and assertions
studiomaxops
Multi-dimensional token-bucket / leaky-bucket rate limiter (per-tenant, per-provider, per-tool)
studiomaxops
Deterministic scheduled-job inspection and removal on the host's own scheduler
Max Meier
Zero-dependency stdio MCP server for Thredz — long-term memory (wiki recall, semantic/keyword search, upsert, reflection signals), durable goals & tasks, and agent-to-agent messaging over the Thredz REST API. Point any Model Context Protocol client at it
...more