Vexp
I built vexp because AI coding agents have two expensive problems: they waste tokens reading irrelevant code, and they forget everything between sessions. The token problem: agents read entire files linearly to build context. On a medium TypeScript project, a single query was consuming ~18k tokens — most of it irrelevant. vexp builds a dependency graph from the AST (who calls what, who imports what, what types flow where) and serves only the relevant subgraph as a token-budgeted capsule. ~2.4k tokens instead of ~18k, with better response quality because the context is precise. The memory…
Vexp launched on March 31, 2026 with 14 votes, #999 of 23,380 launches that month and more than 90% of that year's launches.
What it does
vexp gives your AI coding agent only the code that matters. It indexes your codebase into a dependency graph locally (Rust, tree-sitter, SQLite) and serves ranked context via MCP. Instead of reading entire files, the agent gets the relevant functions in full and connected symbols as skeletonized signatures. Works with Claude Code, Cursor, Copilot, and 9 more agents. 30 languages. Session memory with staleness detection. Zero cloud, zero network calls. Your code never leaves your machine.
Local-first reliability layer for AI coding agents: a context engine that feeds your agent only the code that matters, plus mechanical verification of the work when it says it is done. No cloud, no account, no code leaving your laptop. 65-70% token reduction.
The right context before your agent writes a line, and a mechanical check before it says done. Up to 70% fewer tokens. 33,000+ installs · 14 agents · 34 languages “If you are working on enterprise software with complex, interconnected modules, you already know the pain of token bloat. Vexp does exactly what it promises: it actually maps out the internal calls and dependencies between modules and drastically reduces API bills.” Every session, the agent scans files, guesses dependencies, and floods its context with code it never uses. Vexp gives it precise structure instead. Three steps in, one check out: when the agent declares the work done, Horizon verifies it mechanically…from vexp.dev
In the maker’s words, at launch
I built vexp because AI coding agents have two expensive problems: they waste tokens reading irrelevant code, and they forget everything between sessions. The token problem: agents read entire files linearly to build context. On a medium TypeScript project, a single query was consuming ~18k tokens — most of it…
Does a similar job
all alternatives →
Context GatewayMar 2026 · github.com · ▲220We built an open-source proxy that sits between coding agents (Claude Code, OpenClaw, etc.) and the LLM, compressing tool outputs before they enter the context window. Demo: https://www.youtube.com/watch?v=-vFZ6MPrwjw#t=9s. Motivation: Agents are terrible at managing context. A single file read or grep can dump thousands of tokens into the window, most of it noise. This isn't just expensive — it actively degrades quality. Long-context benchmarks consistently show steep accuracy drops as context grows (OpenAI's GPT-5.4 eval goes from 97.2% at 32k to 36.6% at 1M…
AgentmemoryMay 2026 · agent-memory.dev · ▲310Persistent memory for Claude Code, Codex & coding agents
LangAlpha – what if Claude Code was built for Wall Street?Apr 2026 · github.com · ▲148Some technical context on what we ran into building this. MCP tools don't really work for financial data at scale. One tool call for five years of daily prices dumps tens of thousands of tokens into the context window. And data vendors pack dozens of tools into a single MCP server, schemas alone can eat 50k+ tokens before the agent does anything useful. So we auto-generate typed Python modules from the MCP schemas at workspace init and upload them into the sandbox. The agent just imports them like a normal library. Only a one-line summary per server stays in the prompt. We have around 80…
Sx – an open-source package manager for AI skills, MCPs, and commandsMay 2026 · github.com · ▲50Claude Code is amazing for individuals but terrible for teams. I spent the last six months building my perfect mix of MCP servers, subagents, commands and skills. Claude Code writes 100% of the code now, but sharing this with my team? Painful. 1. Blog posts and workshops don't work (and don't scale) 2. Committing tools to N repos means N PRs. A tool changes? N more PRs. 3. Same tools, different files or folders for each AI assistant; not everyone uses Claude Code It is like 1999 all over again; FTPing files up to prod and versioning with ".bak" suffixes. So I built sx. It is your team's…
OpenBrowser-AIApr 2026 · openbrowser.me · ▲76Your AI agent is burning 6x more tokens than it needs to just to browse the web. We built OpenBrowser MCP to fix that. Most browser MCPs give the LLM dozens of tools: click, scroll, type, extract, navigate. Each call dumps the entire page accessibility tree into the context window. One Wikipedia page? 124K+ tokens. Every. Single. Call. OpenBrowser works differently. It exposes one tool. Your agent writes Python code, and OpenBrowser executes it in a persistent runtime with full browser access. The agent controls what comes back. No bloated page dumps. No wasted tokens. Just the data your…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 20d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 28d ago · world.pixverse.video
Source: Product Hunt launch ↗ · launch text quoted in part
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
Littlebird▲709Littlebird remembers everything across your meetings, messages, docs, and apps, then takes action so you can recall any detail and never lose context.
AI · Mar 2026 · littlebird.ai
