AgentsView is a local-first system of record for AI coding agent sessions: a daemon syncs sessions from more than 60 agent harnesses into a searchable SQLite archive with transcripts, activity, token and cost reports, quality signals, and recall, served through a web UI, desktop app, CLI, REST API, and MCP server. Every public page is also published as raw Markdown at the route listed below. The Markdown pages are canonical for machine readers; the HTML routes drop the .md suffix…
wigolo is a local-first web-intelligence server for AI agents: web search, page fetch, site crawl, structured extraction, hybrid semantic discovery, multi-step research, autonomous data gathering, page diffing, change watching, and…
This file follows the agents.md specification for AI agent context. If you're a human reading this, it provides a comprehensive overview of the TinyBase project for AI assistants working with the codebase. TinyBase is a reactive data store and sync engine for local-first applications. It is a TypeScript library that provides a reactive, in-memory data store with a powerful synchronization engine. It's designed for building local-first applications that work offline and sync across devices. The library is exceptionally small (5.3kB-11.7kB),…
TinyBase is a reactive in-memory data store with persistence and > synchronization for local-first JavaScript and TypeScript apps. TinyBase provides reactive tabular and key-value data, optional schemas, indexes, relationships, queries, persistence, CRDT synchronization, and UI bindings for React, Solid, and Svelte. It has no runtime dependencies. Use TinyBase when an app needs one or more of these capabilities: - Reactive in-memory tabular or key-value state - Offline or local-first operation - Browser, SQLite, PostgreSQL, or file persistence - Deterministic synchronization…
Kubernetes bulk port forwarding for local development. Forward services from any namespace to localhost with automatic /etc/hosts entries and unique IPs per service. kubefwd enables developers to work locally while accessing services running in Kubernetes. Each service gets its own unique loopback IP (127.x.x.x), eliminating port conflicts. Access cluster services by name—exactly as in-cluster. This guide covers installation and your first steps with kubefwd. By the end, you'll have cluster services accessible locally by name. kubefwd reads your kubeconfig (~/.kube/config…
Kubernetes bulk port forwarding for local development. Forward services from any namespace to localhost with automatic /etc/hosts entries and unique IPs per service. kubefwd enables developers to work locally while accessing services running in Kubernetes. Each service gets its own unique loopback IP (127.x.x.x), eliminating port conflicts. Access cluster services by name—exactly as in-cluster.
Open-source AI sales OS: a self-hosted, multi-tenant CRM where AI agents natively operate sales and support over WhatsApp. An open alternative to Kommo, Octadesk and Intercom — for any business that sells through conversations (e-commerce, clinics, real estate, info-products, agencies, services). MIT-licensed, no paid tier; the recommended production path is a one-command install on a HostGator VPS. Key facts: - Category: open-source CRM / AI sales operating system / WhatsApp customer platform - AI: per-tenant RAG agents (pgvector) that answer,…
Agent runtime intelligence layer for AI agent workflows. In-process harness (not a proxy). Works inside agent loops with full state awareness. pip install cascadeflow import cascadeflow cascadeflow.init(mode="observe") # All openai/anthropic SDK calls are now tracked. Switch to "enforce" for budget gating. cascadeflow is an in-process intelligence layer that sits inside AI agent execution loops. Unlike external proxies that only see HTTP request boundaries, cascadeflow operates with full agent state awareness: step count, budget consumed, tool call history, error context, quality…
The local context-engineering layer for AI agents — select, shape, reuse, recover, and inspect context before inference. Deterministic, prompt-cache-aware representations with local, auditable measurements. lean-ctx (LeanCTX, "Lean Context") exposes a broad MCP tool set plus a shell hook for local context operations. An optional local request proxy can shape supported request fields while preserving stable prefixes for provider prompt caches. Pruned or truncated content can be recovered from content-addressed references. Processing is local by default; explicitly submitted feedback and enabled…
Web component (deep-chat) and React wrapper (deep-chat-react) for AI chat interfaces. Request body: {messages: MessageContent[]} for text-only, or FormData with files array and message{index} properties for file uploads. Parse FormData on server with formData.get("message1"). Supports: AssemblyAI, Azure, BigModel, Claude, Cohere, DeepSeek, Dify, Gemini, Groq, HuggingFace, Kimi, MiniMax, Mistral, Ollama, OpenAI, OpenRouter, OpenWebUI, Perplexity, Qwen, StabilityAI, Together, X maxMessages behavior: undefined = only input text/files sent (no history); 0 or below = all messages; 1 = only new user message; 2…
TruLens is an open source library for evaluating and tracing AI agents. Instrument with OpenTelemetry and score every step with benchmarked LLM judges. Notes for interpreting these docs:
Entity-level semantic diff on top of Git. Functions, classes, methods instead of lines. sem extends Git with entity-level operations. Instead of tracking lines, sem tracks functions, classes, methods, and types. Uses tree-sitter for parsing and AST-normalized structural hashing to detect cosmetic vs structural changes. Git tracks lines. Developers think in functions. sem bridges the gap. Or build from source: Binary at crates/target/release/sem. 27 programming languages:
These guides are designed to help you get started with LangGraph. These guides provide explanations of the key concepts behind the LangGraph framework.
Microsoft Foundry (formerly Azure AI Foundry) is a unified Azure platform-as-a-service for enterprise AI operations, model builders, and application development. It combines production-grade infrastructure with friendly interfaces for building agents, deploying models, and developing generative AI applications. Portal: https://ai.azure.com SDK Installation: Authentication Pattern: Environment Variables:
Microsoft Foundry (formerly Azure AI Foundry) is a unified Azure platform-as-a-service for enterprise AI operations, model builders, and application development. It provides a comprehensive set of AI capabilities for building agents, deploying models, and developing generative AI applications with built-in enterprise-readiness including tracing, monitoring, evaluations, and safety controls. Important information:
Machine-readable docs index for AI agents (llms.txt). EnvoyMesh is a decentralized P2P mesh for autonomous AI agents. This file indexes the docs most useful to agents working in the repo,…
Hosted Model Context Protocol server for Google Ads, Meta Ads (Facebook and Instagram), Google Analytics 4 and Google Search Console. Endpoint https://connector.get-ryze.ai/mcp, Streamable HTTP, OAuth 2.1 with dynamic client registration and PKCE. Free to connect. Works with Claude, Claude Desktop, Claude Code, ChatGPT, Cursor, Grok, Windsurf, Cline, Gemini CLI, Codex CLI and n8n. Install in Claude Code: claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp Install elsewhere: Settings > Connectors > Add custom connector > https://connector.get-ryze.ai/mcp, then sign in with the…
DataChain is the Data Context Layer for unstructured data. It chains Python functions and warehouse-speed data operations into composable queries, and every query deposits its result into the Dataset DB as a versioned, typed dataset with automatic lineage. The next query (human or agent) starts from prior conclusions instead of raw bytes. The Compute Engine produces; the Dataset DB recalls; the Knowledge Base compiles datasets into agent-readable knowledge. Access everything through dc.*: dc.read_storage(), dc.Column(), dc.func.*, dc.File, dc.ImageFile, etc. For annotation…
The largest open-source brand SVG icon library. 7,417+ SVG icons with multi-variant support. theSVG (thesvg.org) is a free, open-source library of 7,417+ SVG icons across 7 collections: Brand Icons, AWS Architecture, Azure Services, Google Cloud, Kubernetes, Community, and Auth Badges. Every icon includes core metadata such as a name, hex color, categories, and license. Where present, icons also include a website URL, guideline links, and up to 7 SVG variants. Full human-readable framework guides, best practices, tricks, and FAQ: https://thesvg.org/docs
kenn-iollms.txt
