LightfieldAI-native CRM that builds itself and does work for you
Promoted
Launching today
Meet Contextual
A local-first temporal context engine for your AI agents
1 follower
A local-first temporal context engine for your AI agents
1 follower
Contextual is a local-first CLI/MCP daemon that gives any AI coding agent (Claude Code, Cursor, Copilot, and more) a persistent, structurally accurate memory of your codebase built from a real dependency graph and AST parsing, not text similarity or guesswork. 0 bytes of your code ever leave your machine. 5 mins to contextualise on 50k LOC, 94.5 ms median recall latency, so your AI tools finally know your code as well as you do.
Free Options
Launch Team / Built With
Hi everyone — excited to bring Contextual to Product Hunt today.
Before this, I was freelancing on enterprise agentic AI and ML prediction
systems. The same problem kept showing up: once an agent hit a real
production codebase, it would confidently guess wrong about code it didn't
understand, and I'd lose hours rewriting the same prompt just to get back
to where I already was.
Contextual is a local daemon that gives any MCP-capable AI client (Claude
Code, Cursor, Copilot, and more) a real, persistent memory of your
codebase — built from an actual dependency graph and AST parsing, not text
similarity.
Honest numbers: 5 mins to contextualise 50k LOC, 94.5ms median recall latency,
0 bytes of your code ever leave your machine. Median Recall@10 across 11
real repos is 85%, with a true range of 54–98% depending on the
codebase.
It's $10/mo, proprietary, with a full-featured 14-day trial rather than a
capped free tier — capping usage would mean tracking what you do locally,
and the whole point is that we don't.
Built this to help developers ship faster. Would love your honest take,
especially anything that surprises or breaks for you — thanks for
checking it out!