Remember

Lint-AI

Lint for AI-generated knowledge

Drop Lint-AI into your coding agent as its memory: retrieval-grade recall over past sessions and notes, plus corpus linting that flags orphaned pages and missing cross-references — instead of just storing everything it sees.

Claude CodeCodexGemini CLIAntigravity CLI (AGY)

Open source · Apache-2.0 · Rust · runs entirely on your own hardware

lint-ai

$ lint-ai ./docs

Issues:

- Orphan page: quickstart.md

- Missing cross-ref in claude-code-hooks-design.md -> [[claude code]] (high)

- Missing cross-ref in claude-code-hooks-design.md -> [[codex]] (high)

- Low link density in tier1-light-understanding.md (outgoing 0, avg 0.1)

Stats:

pages: 31

top concepts: lexical data (9), codex (6), claude code (5)

86.9%
Recall@5
LongMemEval-S, 500 questions
87.1%
MRR
LongMemEval-S, 500 questions
5.1ms
Avg. query latency
single CPU core, no GPU

What it saves your agent

Diagnostic smoke-run numbers from a single repetition — a direction, not a guarantee. See the full run conditions.

ArmTimeInput tokensRecall
Claude, native memory22.47s123,5362/3
Claude, Lint-AI only7.05s15,9143/3
Codex, native memory38.13s151,0002/3
Codex, Lint-AI only18.53s62,4782/3

What it does

Lexical + graph indexing

Entity and term tables alongside a graph structure for links and symbols, so retrieval understands structure, not just text.

Corpus-level linting

Flags orphan pages, missing cross-references, and low link density across your whole document set — not just one file at a time.

Works as memory in your CLI

Plugs into Claude Code, Codex, Gemini CLI, and Antigravity CLI (AGY) as their memory, via lifecycle hooks and an MCP search interface — no manual tool selection required.

Semantic + LLM-ready queries

Query modes include semantic retrieval and context generation formatted for direct use in an LLM prompt.

Configurable scope

A single lint-ai.json controls filtering and scoping, so you decide what gets indexed and what gets ignored.

CPU-only by design

Built in Rust with an optional spaCy NER integration. No GPU in the critical path.

Get started

# clone and build from source

git clone https://github.com/RooAGI/Lint-AI/

cd Lint-AI

cargo build --release

Available as a CLI or a library. See the README for provider integration flags and lint-ai.json configuration, or the integration benchmark methodology for how memory performance is measured across each CLI.

Let's talk about your agents

Whether it's agent memory, a no-code build, or in-house model serving, tell us what you're working on and we'll point you to the right place to start.

Free Architecture Review

We'll look at your current agent setup and flag where memory, tooling, or serving is the bottleneck

Custom Deployment Guidance

Tailored recommendations for your use case, team size, and hardware

Enterprise Support Options

24/7 support, priority updates, and dedicated engineering resources

Built for teams running agents in production

From open-source Lint-AI users to enterprises deploying custom inference — RooAGI covers the whole stack.

Request Information