|
React Recall
React Recall captures every click, log, and network request—so your coding agent can find bugs in seconds.
|
79 % |
|
Recall — Local memory for your coding agents
Recall indexes your Claude Code, Codex, OpenCode, and Pi sessions into a local SQLite knowledge base so you can search, resume, and build on everything you h...
|
78 % |
|
WDIMTM — What Does It Mean To Me?
Browser-native AI for the moment you see something on the web and think: what does this mean to me? Select text. Stay on the page. Understand.
|
76 % |
|
TheContextCache — Your repo remembers why
Capture the why from your AI chats once — decisions, findings, notes — and recall it in any AI tool.
|
76 % |
|
PromptCacheAI | Cache Repeated LLM Requests Before They Cost You
PromptCacheAI is an application-layer LLM cache that checks exact and semantically similar prompts before your app calls OpenAI, Claude, Gemini, or custom mo...
|
76 % |
|
ogrep v0.12.0 — semantic code search for AI agents and humans
ogrep v0.12.0 — semantic code search for AI agents and humans. Hybrid fulltext/vector search, MCP-native, SQLite-backed, local or cloud embeddings. Auto-refr...
|
76 % |
|
fantazm.ai · open-source modular AI tooling
A small lab building local-first, security-first, open-source AI tooling. Composable parts — an agent runtime, a sandbox, an egress proxy, memory, identity,...
|
76 % |
|
Recalld: The memory layer for AI Agents
Recalld is the memory layer for AI agents. Recall, not retrain. Persistent, curated memory that returns only what answers the question. 88.7% on the LoCoMo b...
|
76 % |
|
inra — the recall layer for AI agents
Drop-in inference-time memory for AI agents. Store what happened, recall what matters, at query time. Built for production.
|
76 % |
|
mll-proxy | Semantic Cache for LLM APIs
Semantic caching proxy for LLM APIs. Cache responses by meaning, not by URL. Drop-in OpenAI-compatible, Rust-powered.
|
75 % |
|
ZeCache — Semantic Caching for LLM Applications
ZeCache is a managed semantic caching API that helps AI applications avoid unnecessary LLM calls by safely reusing equivalent responses.
|
75 % |
|
ManagePrompt — Local LLM Call Debugger
Capture every LLM API call during development. Full request/response details, token usage, cost estimation, and latency — all in a local web UI.
|
75 % |
|
PostgreSQL Configuration API
A basic JSON API for generating a PostgreSQL configuration. Specifically built for use with configuration management systems.
|
75 % |
|
RAGrep — Local Semantic Recall
|
75 % |
|
Intentgine — Cached Intent-as-a-Service
Turn natural language into structured tool calls with semantic caching and built-in learning.
|
75 % |
|
BeigeBox — LLM Middleware Control Plane
BeigeBox: OpenAI-compatible LLM middleware with 4-tier intelligent routing, semantic response caching, WASM post-processing, multi-model harness, and local-f...
|
75 % |
|
Remnis
Context engine for macOS
|
75 % |
|
ziflux: SWR caching for Angular resource()
Zero-dependency, signal-native stale-while-revalidate caching for Angular resource().
|
75 % |
|
CacheCore: Stop paying for the same LLM call twice.
CacheCore sits between your agents and the LLM API. Duplicate and near-identical calls return from cache in milliseconds. One config change, no rewrites.
|
75 % |
|
MemDev — Your terminal has a memory. Finally.
MemDev captures your terminal sessions, compresses them with CodeT5-Large, and injects relevant context into Claude Code, Cursor, or any AI assistant.
|
75 % |
Detected social media networks