https://github.com/reaatech/agents-md-kit

Linter, validator, and examples for AGENTS.md and SKILL.md — durable agent instruction artifacts. CLI to lint, validate schema, and scaffold new files. Curated examples directory.

agentic-ai agents-md ai-agents claude-code cli codex coding-agents cursor developer-tools instruction-artifacts linter opencode scaffold schema-validation skills-md typescript validator windsurf

Last synced: 16 Jul 2026

https://github.com/reaatech/a2a-reference-ts

Enterprise-grade TypeScript implementation of Google's Agent-to-Agent (A2A) protocol with a bidirectional A2A ↔ MCP bridge, OAuth2/JWT auth, Redis/Postgres persistence, SSE streaming, and OpenTelemetry observability.

a2a agent-orchestration agent-to-agent agentic-ai ai-agents express google-a2a hono llm-tools mcp mcp-bridge model-context-protocol monorepo multi-agent oauth2 opentelemetry postgres redis sse typescript

Last synced: 28 May 2026

https://github.com/reaatech/prompt-injection-bench

Reproducible benchmark and test corpus for prompt-injection defenses — adapters for common sanitization libraries, scored leaderboard, CI-ready. Security-adjacent, evergreen.

adversarial-testing agentic-ai ai-safety ai-security benchmark defense llm llm-security prompt-injection red-teaming sanitization security test-corpus typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-server-doctor

Lighthouse for MCP servers — CLI diagnostic and profiling tool that runs 8 checks against any MCP endpoint and produces a letter-graded report card with latency profiling, concurrency stress testing, schema validation, and CI/CD integration.

ai-agents cicd cli developer-tools devops diagnostics health-check json-rpc latency load-testing mcp mcp-tools model-context-protocol nodejs npm-package observability opentelemetry profiling testing typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/llm-cost-telemetry

Small library wrapping provider SDKs to emit per-call cost spans, aggregate by tenant/feature/route, and export to CloudWatch, Cloud Monitoring, and Phoenix. Tiny scope, universal need.

agentic-ai anthropic arize-phoenix bedrock cloud-monitoring cloudwatch cost-tracking finops gemini llm-cost llmops observability openai opentelemetry telemetry typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/voice-agent-kit

Real-time voice agent pipeline — Twilio to STT to MCP agent with vector retrieval to TTS to Twilio. Latency budget enforcement, barge-in handling, session continuity, Cal.com tool integration. Few clean public references exist for this.

agentic-ai calcom conversational-ai latency-optimization rag real-time speech-to-text stt telephony test-to-speech tts twillio typescript vector-search voice-agent voice-ai

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-replay

Record and deterministically replay agent interactions for debugging. Capture LLM calls, tool use, and routing decisions as trace files. Replay with stubs or live. Diff outputs across versions.

agentic-ai ai ai-agents developer-tools llm mcp observability opentelemetry replay testing typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-catalog

Registry server for MCP server discovery across teams. Register, search, browse, and health-check organizational MCP servers. Agents discover tools dynamically via the catalog's own MCP interface.

agentic-ai ai ai-agents developer-tools mcp mcp-server mcp-tools model-context-protocol service-discovery typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/session-continuity-kit

Multi-turn session management for agent systems. Conversation history windowing, token budget management, context compression, and persistence. Adapters for Firestore, DynamoDB, Redis.

agentic-ai ai ai-agents developer-tools llm mcp rag typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-auth-proxy

Identity-aware proxy for agent-to-service communication. OAuth2 token management, per-user credential isolation, scope enforcement, and audit logging. Agents act on behalf of users safely.

agent-auth agent-proxy agentic-ai ai ai-agents developer-tools mcp oauth2 security typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/invoicing-app

Free Desktop Invoicing App for Contractors and Freelancers

electron electron-app invoice invoice-generator invoice-management invoice-pdf invoicing

Last synced: 16 Jul 2026

https://github.com/reaatech/terraform-mcp-gcp-cloudrun

Drop-in Terraform module to deploy an MCP server on GCP Cloud Run with Firestore, Secret Manager, and Pub/Sub. Production-ready defaults, multi-environment, OTel wired.

agentic-ai cloud-run deployment devops firestore gcp google-cloud hcl iac infrastructure-as-code mcp opentelemetry pub-sub secret-manager serverless terraform

Last synced: 16 Jul 2026

https://github.com/reaatech/postmill-app

AI-native social media post management platform for 36+ channels — governed multi-provider AI (26 providers) for on-brand content, smart replies, compliance guardrails & spend caps, plus persisted analytics, a cross-channel comment inbox, campaigns, and native polls.

ai ai-agents analytics generative-ai langchain llm nestjs nextjs openai postgresql postiz prisma rag scheduling self-hosted social-media social-media-management social-media-scheduler temporal typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/circuit-breaker-agents

Circuit breaker library designed for agent-to-tool and agent-to-agent communication. Per-tool breakers, confidence-aware tripping, cost-aware tripping, gradual HALF_OPEN recovery, and state persistence.

agentic-ai ai ai-agents circuit-breaker developer-tools llm mcp

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-memory

Long-term memory layer for agents across sessions. Memory extraction, semantic retrieval, fact storage, decay, and contradiction resolution. Adapters for Qdrant, Pinecone, and Postgres pgvector.

agentic-ai ai ai-agents embeddings generative-ai llm rag semantic-search typescript vector-database

Last synced: 16 Jul 2026

https://github.com/reaatech/classifier-evals

Offline classifier evaluation harness — dataset loader, confusion matrices, LLM-as-judge with cost accounting, regression gates for CI, Phoenix/Langfuse exporters. Built for intent classifiers but works on any classification task.

agentic-ai arize-phoenix ci-cd classifier confusion-matrix evaluation-harness intent-classification langfuse llm-as-judge llm-eval mlops observability regression-testing testing-tools typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/structured-output-repair

Catch and fix malformed LLM structured outputs instead of crashing. Strip fences, fix JSON syntax, coerce types, re-prompt if unrepairable. Zod schema in, reliable parsed output out.

agentic-ai ai ai-agents anthropic developer-tools json llm openai typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-load-test

Load testing framework purpose-built for MCP servers. Realistic tool call sequences, concurrent multi-turn sessions, transport-aware profiling, and latency histograms per tool. Not wrk — MCP-native.

agentic-ai ai cli developer-tools mcp mcp-server mcp-tools model-context-protocol testing typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-handoff-protocol

Transfer conversations between agents mid-session. Handoff triggers, context packaging, history compression, routing, acceptance confirmation. Works over MCP and A2A.

agentic-ai ai ai-agents developer-tools llm mcp model-context-protocol multi-agent typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/secret-rotation-kit

Zero-downtime multi-key secret rotation. Overlapping validity windows, propagation verification, and revocation. Adapters for AWS Secrets Manager, GCP Secret Manager, and Vault.

aws developer-tools devops docker security typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-runbook-generator

CLI that ingests a service repo and generates a draft operator runbook (alerts, dashboards, failure modes, rollback steps) using an agent. Directly evidences the operator-package deliverable pattern from Fortune-10 engagements.

agentic-ai ai-agents alerts automation cli code-analysis devops documentation incident-response observability runbook-generators sre typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-gateway

Production MCP gateway — OAuth/OIDC + API key auth, per-tenant rate limiting, schema enforcement, tool allowlists, request/response logging, audit trail, fan-out to multiple upstream MCP servers, response caching. The Kong/Envoy for MCP.

agentic-ai api-gateway audit-log auth caching fan-out mcp mcp-gateway mcp-server multi-tenant oauth oidc proxy rate-limiting reverse-proxy schema-validation security typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/guardrail-chain

Composable input/output guardrail pipeline for agent systems. Chain PII redaction, prompt injection detection, topic boundaries, and output filters with short-circuit logic and latency budgets.

agentic-ai ai ai-agents developer-tools generative-ai llm mcp security typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-server-starter-ts

Production-grade MCP server template — StreamableHTTP + SSE transports, Zod schemas, OTel instrumentation, idempotency middleware, prompt-injection sanitization, structured logging, Jest harness, Terraform for Cloud Run + AWS.

agentic-ai aws boilerplate cloud-run idempotency jest llm mcp mcp-server model-context-protocol observability opentelemetry prompt-injection sse starter streamable-http template terraform typescript zod

Last synced: 16 Jul 2026

https://github.com/reaatech/llm-cache

Semantic caching layer for LLM calls. Exact-match and embedding-similarity caching with model-version-aware invalidation, use-case segmentation, and cost-saved tracking. Adapters for Redis and DynamoDB.

agentic-ai ai ai-agents anthropic embeddings llm llm-cache openai semantic-search typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/sl-skills

Agent-consumable skills for the REAA Standard Library — auto-generated from the live catalog at reaatech.com

Last synced: 16 Jul 2026

https://github.com/reaatech/tool-use-firewall

Policy enforcement proxy between agents and MCP tools. Cost caps, rate limits, argument validation, human-in-the-loop approval for destructive ops, read-only mode, and full audit logging. MCP-native.

agentic-ai ai ai-agents developer-tools llm mcp mcp-gateway mcp-server model-context-protocol security typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/agent-eval-harness

End-to-end agent evaluation — trajectory eval, tool-use correctness, cost-per-task, latency budgets, regression suites with golden trajectories, LLM-as-judge with calibration. For full agent runs, not just classifiers.

agent-evaluation agentic-ai ai-agents cost-tracking latency llm-as-a-judge llm-as-judge llm-eval mlops observability regression-testing testing-tools tool-use trajectory-evaluation typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/funcdock

Self-hosted Node.js FaaS platform — run dozens of functions in a single container with zero cold starts, hot reload, and automatic MCP tool exposure for AI agents. Built-in React dashboard, OpenTelemetry tracing, TypeScript native, shared layers, cron scheduling, and CI/CD deployments.

ai-agents cicd cron-jobs dashboard developer-tools devops docker express faas functions-as-a-service hot-reload llm-tools mcp microservices model-context-protocol nodejs opentelemetry self-hosted serverless

Last synced: 16 Jul 2026

https://github.com/reaatech/faas-hot-runtime

Minimal open-source FaaS runtime on EKS — warm pod pool, sub-100ms invoke, HTTP + queue triggers, OTel built in. Lambda without cold starts. Based on the pattern from a production agentic AI platform.

agentic-ai aws cold-start container-orchestration eks faas function-as-a-service kubernetes lambda-alternative microservices opentelemetry runtime serverless typescript warm-pods

Last synced: 16 Jul 2026

https://github.com/reaatech/idempotency-middleware

Framework-agnostic idempotency cache middleware for TypeScript. Handles concurrent duplicates, partial failures, and TTL. Adapters for Firestore, DynamoDB, Redis, and in-memory.

aws developer-tools docker typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/multi-tenant-mcp

Multi-tenancy primitives for MCP servers. Tenant identification from auth, per-tenant rate limits, tool visibility, cost accounting, artifact isolation, and config scoping. Middleware, not a framework.

agentic-ai ai ai-agents llm mcp mcp-server model-context-protocol typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/reaatech

Last synced: 16 Jul 2026

https://github.com/reaatech/bicycle-brands-models

Great bicycle data, freely shared. A structured dataset of brands, models, and sizes—so you don’t have to hunt for it like we did. Enjoy!

bicycle-infrastructure bicycle-model bicycles bikes

Last synced: 16 Jul 2026

https://github.com/reaatech/prompt-version-control

Git-like versioning for prompts with eval-gated promotion. Store, diff, tag (draft/staging/production), A/B serve, and gate promotion on eval results. Self-hosted, not SaaS.

agentic-ai ai ai-agents developer-tools generative-ai llm mcp mcp-server prompt-engineering typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/llm-judge-toolkit

Calibrated LLM-as-judge library. Multi-judge consensus, position bias detection, calibration against human labels, cost tracking, and caching. Judge prompts for faithfulness, relevance, coherence, and safety.

agentic-ai ai ai-agents anthropic developer-tools generative-ai llm openai typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/mcp-contract-kit

Conformance test suite for MCP servers — registry compliance, protocol conformance, and routing correctness verification. Point it at any MCP endpoint to validate spec adherence.

agentic-ai conformance-testing contract-testing developer-tools mcp mcp-server model-context-protocol protocol-validation spec-compliance test-suite testing-tools typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/confidence-router

Decision engine for route/clarify/fallback patterns. Confidence-gated intent routing with configurable thresholds, clarification prompt generation, pluggable classifiers, and eval harness integration.

agentic-ai ai ai-agents developer-tools generative-ai llm mcp typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/context-window-planner

Optimal context window packing as a resource allocation problem. Fit system prompts, history, RAG chunks, tool schemas, and generation buffer into a token budget with priority-based strategies.

agentic-ai ai ai-agents anthropic developer-tools llm openai prompt-engineering rag typescript

Last synced: 16 Jul 2026

https://github.com/reaatech/webhook-relay-mcp

MCP server that receives webhooks from external services and translates them into agent-consumable events. Signature validation, payload normalization, and tools for subscribe/poll/history.

agentic-ai ai ai-agents mcp mcp-relay mcp-server mcp-tools model-context-protocol typescript webhooks

Last synced: 04 Jun 2026

https://github.com/reaatech/agent-mesh

Reference multi-agent orchestrator — stateless design, YAML agent registry with SIGHUP hot-reload, confidence-gated route/clarify/fallback, per-agent circuit breakers (Firestore-persisted), session-bypass middleware, three-layer contract tests.

agent-framework agentic-ai circuit-breaker cloud-run confidence-gating contract-testing distributed-systems firestore hot-reload llm mcp multi-agent-orchestration orchestrator session-management stateless typescript yaml-config

Last synced: 03 Jul 2026

https://github.com/reaatech/llm-router

Cost/latency/judgment-based model router — pluggable routing strategies, fallback chains, per-model cost telemetry, OTel spans, eval hooks. Ships a sample config for the frontier-judge + code-workhorse + local-inference pattern.

agentic-ai ai-infrastructure anthropic bedrock cost-optimization fallback gemini latency llm llm-router llmops model-routing multi-model ollama openai opentelemetry routing-engine typescript vllm

Last synced: 16 Jul 2026