|
Poggle | Shared memory for coding agents
Poggle is shared memory for coding agents. Agents log what they do as they do it, so the next one — different tool, different account, different machine — pi...
|
84 % |
|
Atlas Scout - Token-Efficient Code Navigation MCP Server for AI Agents
A local, read-only MCP server that gives Claude Code, Codex, and Cursor exact symbols, source ranges, and references in one call - so your agent stops burnin...
|
82 % |
|
ctxpilot — Memory engine for AI coding agents
Cut repeated AI coding context with a local memory engine, token analytics, and Claude Code beta hooks.
|
81 % |
|
greplm — Code intelligence that keeps AI agents on budget
greplm gives AI coding agents instant, precise answers from your codebase — using up to 99% fewer tokens. Works with Cursor, Claude Code, Copilot, and any MC...
|
81 % |
|
Gesta — Observability, audit & security for coding agents
Gesta measures, governs, and guides every coding-agent session — on the machine where it happens. See what agents authored, audit who wrote what, hold the da...
|
81 % |
|
Tokeasy — Cut your AI coding-agent token bill
Cut your AI coding-agent token bill. Ten drop-in MCP tools for Claude Code, Codex, Copilot, Cursor & OpenClaw — fewer calls, smaller context, same model. On...
|
81 % |
|
agent-wrapped — your year in AI coding agents
Turn your local Claude Code and Codex logs into a stats card for your GitHub README.
|
81 % |
|
IntelOps · Token FinOps for Agentic AI
Token FinOps for agentic AI builders. See where your agents burn tokens, compare models, cut spend by 10x.
|
80 % |
|
SuperLeanAI — Prompt Compression & Token Efficiency Proxy
SuperLeanAI is a prompt compression and context compression proxy for AI coding agents: it debloats verbose tool definitions, MCP server schemas, and Skills...
|
80 % |
|
Cards — lasting memory for your AI agents
Your AI forgets your project every session — what you decided, what you ruled out, what’s still open. Cards gives your project a memory that lasts, and your...
|
80 % |
|
Persistent memory for AI agents that cuts token usage · Tulimoa
Tulimoa Memory gives your AI agent a persistent memory over MCP. It stores what happens, sorts it into a knowledge graph overnight and hands the right contex...
|
80 % |
|
TokenSentinel — catch AI token waste mid-run
Catch AI token waste mid-run—before the bill. For AI agents in your app and in coding tools.
|
80 % |
|
nogrep — code context for AI agents
nogrep turns a repository into a code graph and hands AI agents the relevant slice. 14K tokens per bug-fix session instead of 220K.
|
80 % |
|
stuff.md — tokenmaxing for agent tools
A research notebook for tokenmaxing: building agent tools (CLI / MCP / code-exec) that finish a task in the fewest tokens — and a fair benchmark to prove whi...
|
80 % |
|
TokenDam — find where your LLM agent burns tokens
Paste your agent's trace. See exactly where you're burning tokens — and how to stop. Runs 100% in your browser — nothing you paste is uploaded.
|
80 % |
|
Memmo — Long-term memory for AI coding agents
Memmo is operational memory for AI coding agents. It turns your repo, agent sessions, docs, and past mistakes into living context — so Claude Code walks in a...
|
80 % |
|
knoot: multiplayer for AI coding agents
knoot makes coding agents multiplayer. Every agent on a team sees what the others are doing, inherits what they have learned, and is stopped before two of th...
|
80 % |
|
Claude Agent Harness — docs, checks and hooks for coding agents
One command installs rules, a docs kit, checks and session hooks into your repository. Coding agents keep context, stay in their zones and stop overpaying to...
|
80 % |
|
Covector · ctx-loop — living context for AI coding agents
Give your agent context that stays true. ctx-loop gives you — and your team — living context: small, sourced, expiring cards your agent pulls in fresh, every...
|
79 % |
|
Tokenade — cut your AI agent's token bill
Cut the token usage of Claude Code, Cursor, Codex and Copilot. Ranked #1 on the Token-Harness Optimizer Leaderboard. Zero config, one-command install.
|
79 % |
Detected social media networks