Skip to main content
Glama

Leveret

A leveret is a young hare — small, fast, and born with its eyes open.

Leveret is a self-hosted, hybrid engine for private code reviews: the successor to hosted AI review bots for teams that want reviews to run on their own infrastructure. It combines a deterministic static-analysis layer, a code graph built into every checkout, a graded noise filter with durable memory, and adversarial agent contracts — driven by the AI you bring (BYOAI: your provider and model — Anthropic or OpenAI by API key or subscription, or a local OpenAI-compatible endpoint). The engine layer itself never calls an LLM, and Leveret operates no hosted analysis service: reviewed-checkout analysis and provider credentials are never routed through Leveret-operated infrastructure. The model provider you configure does receive the prompts and source excerpts the runner sends it; use a local endpoint to keep even that on your network. An optional webhook relay, if you choose one, sees webhook payloads and a scoped token (see DESIGN.md); self-hosting without it involves no relay.

Priorities are working end-to-end reviews, accurate and actionable reports, consequential defect recall with independent verification and beyond-diff evidence, and honest coverage reporting. The target policy is no arbitrary file-count limits by default, with optional client limits. Today the Java path on main still rejects more than 2,000 files outside its configured roots; #80 removes that default and is not yet merged. Evidence-pack file-detail caps also remain and must be evaluated against coverage quality, not service economics. See DESIGN.md.

How a review works

flowchart TD
    D[/"📄 PR diff"/]:::gh
    S["🔍 scan<br>engines + delta vs base<br>+ profile + memory"]:::core
    R["🐇 review agent<br>five lenses,<br>cross-file blast radius"]:::agent
    V["⚖️ verification agent<br>refute or evidence,<br>three grades"]:::agent
    T[/"📋 tiered findings<br>+ walkthrough report"/]:::gh
    D --> S
    D --> R -- concerns --> V -- survivors only --> T
    S -- bounded post-walk leads --> V
    classDef gh fill:#6ea8fe,stroke:#3d6fd9,color:#111
    classDef tun fill:#ffc86b,stroke:#cc8f22,color:#111
    classDef core fill:#7ed6a2,stroke:#3d9e6a,color:#111
    classDef agent fill:#c9a0f5,stroke:#9059d1,color:#111
    classDef store fill:#9fd8e3,stroke:#4d9aab,color:#111
  1. Deterministic first pass. Engines run only against what the change touches: semgrep (registry security + per-language rulesets, offline-capable), gitleaks (secrets over the commit range), shellcheck, ruff, actionlint, zizmor (workflow security), osv-scanner (lockfile CVEs), typos, jscpd (duplication, corpus-gated), custom semgrep/ast-grep rule packs, and any SARIF-emitting command via profile custom: entries (recipes: psalm taint, hadolint, trivy, …). Delta scanning is on by default with a base ref: findings already present at the base tree are dropped as pre-existing — counted, never silent — with multiset identity (a copy of a known-bad line still surfaces), rename tracking, and surfaced base-pass failures. A code graph is generated into the checkout at the exact reviewed commit, so agents query structure instead of grepping for it.

  2. Accounted filter. After discovery completes, surviving deterministic leads enter verification as one bounded, routed stream. Every supplied concern and lead ends as actionable, priced-noise, false-positive, or dropped; profile and memory suppression happens before routing. Nothing is silent: suppression, exact-mechanism deduplication, overflow IDs/bytes, verifier rationale, and publication state remain separate structured accounting.

  3. Memory that learns from humans. .leveret/memory.jsonl, versioned in the reviewed repo: fingerprint verdicts (optionally anchored to a source line — the memory dies when the line changes) plus conventions — free-text rulings taught by maintainers via learn, injected into the agent prompts as repo case law, able both to suppress noise and to raise findings that violate them.

  4. Adversarial contracts. The discovery walk runs five lenses (correctness and hostile inputs, contract conformance, test honesty, blast radius, and explicitly deferred leads triage) without deterministic lead material, and traces changed symbols to call sites outside the diff. After the walk, the verification agent tries to refute every concern and routed lead; claims it can neither refute nor ground in executed evidence are dropped, not published.

  5. Reporting. Findings publish in importance tiers (critical / major / minor / nit, distinct from engine severity), out-of-diff findings appear with their stated correlation to the change, pre-existing defects adjacent to edited lines return as reminders, and every review carries a walkthrough: per-lens outcomes (clean included), per-file verdicts, post-walk lead/overflow metrics, the engine table, and a run-configuration line naming the harness, model, and thinking level that produced the review.

Related MCP server: reporelay

Ways to run it

GitHub App (autonomous). A self-hosted App layer receives PR webhooks, checks out the head, builds the code graph, runs the scan, drives the standardized runner, and posts the review — inline comments plus walkthrough. The App holds only a GitHub App key and webhook secret; model credentials live exclusively in the runner. Human replies on findings feed learn. Getting started + diagram: docs/app.md.

Standardized runner. leveret-runner-pi drives the review/verify contracts through a pinned Pi runtime. Leveret supplies the system prompt, phase-specific terminal submission schema, and read-only review tools; Pi supplies the provider/model runtime and trusted host resources. Models submit phase results through leveret_submit_phase; assistant text is not parsed as JSON. Host-installed Pi/OMP extensions and hooks, Pi/Claude/ Codex skills, prompt templates, and context are loaded. The reviewed checkout is never Pi's working directory, so its settings, hooks, skills, prompts, MCP configuration, and context cannot extend the session. You choose provider, model, and effort (--model / --effort / --provider, or the matching LEVERET_RUNNER_* env vars; defaults openai/gpt-5.6-sol at high). Every walkthrough records the effective client, model, prompt hash, capabilities, and tool metrics. A custom LEVERET_RUNNER remains the bring-your-own-harness escape hatch.

Autonomous reviews retain a private, owner-controlled audit trace by default: Pi's native per-attempt sessions, normalized harness events, App/scanner/subprocess activity, exact failed output, checksums, and a verified zstd-or-gzip archive under LEVERET_DATA. Raw content never enters default stdout. See Private audit traces for policy, retention, export, security, and inspection controls.

Interactive (MCP). Register the server in any MCP-capable client and drive reviews yourself — the served review/verify prompts arrive with your repo's accumulated rulings substituted in (getting started + diagram: docs/interactive.md):

npm install && npm run build
claude mcp add leveret -- node /path/to/leveret/dist/server.js

MCP tools: scan, ast_search (structural search via ast-grep), context (per-function complexity, churn, recency — prioritization signal, not findings), remember (persist a graded verdict), memory (inspect the store), learn (persist a human-taught convention); MCP prompts: review, verify.

The reviewer toolbelt

The engines and structural indexes are capabilities of the reviewer, not the reviewed repository: install them beside Leveret. Full belt: codegraph, semgrep, gitleaks, shellcheck, ruff, actionlint, zizmor, osv-scanner, typos, jscpd, ast-grep, lizard, and a pre-staged Serena LSP bundle for semantic navigation. From a clone, build one with node dist/runner/prefetch-serena.js --bundle /opt/leveret/serena-bundle and run with LEVERET_SERENA_BUNDLE set to that path (the installed package also exposes leveret-prefetch-serena). Runtime downloads are refused.

Before autonomous model work, Leveret builds and validates exact-checkout CodeGraph index, then warms one Serena symbol query per detected packaged language. Missing indexes fail closed by default; set LEVERET_REQUIRE_INDEXES=0 only for an explicitly degraded run.

Optional Java reference evidence requires the installed Inspect distribution, cache-only artifacts, Linux Bubblewrap and prlimit, and paired LEVERET_INSPECT_JAVA_CONFIG / LEVERET_INSPECT_JAVA_CONFIG_SHA256 host values. It supports one Java module with main/test sources, distinguishes exact overloads and method-reference expressions, and reports unresolved or skipped scope separately. A complete response is not proof a change is safe. See Inspect's supported host and configuration.

npm test        # integration suite; exercises the real tools

Repository layout

This is the canonical Leveret monorepo. Inspect is Leveret's JVM inspection module; it retains an independent Gradle build:

Path

Component

Toolchain

repository root (src/, test/, docs/, …)

Leveret: engine, MCP server, runner, GitHub App

Node.js, TypeScript, npm

inspect/

Inspect: Java classpath and checked method-reference evidence for Leveret

JDK 25+, Gradle wrapper

With a hash-pinned host configuration outside the reviewed checkout, the TypeScript reviewer calls Inspect's isolated JVM worker for exact Java method references outside a pinned diff. Without it, the capability is reported unavailable; structural matches do not become checked calls. Inspect has no separate service or product lifecycle. Its original private history and research are not imported into this public monorepo.

Command

Runs

npm run build / npm test

TypeScript build / test suite (Leveret)

npm run build:inspect / npm run test:inspect

./inspect/gradlew -p inspect build / test

npm run build:all / npm run test:all

Both components, TypeScript first

The JVM build requires JDK 25 or newer. build:inspect and build:all also run the JVM tests, including Linux-only classpath oracles requiring Bubblewrap, Git, Maven, JDK 21, and network access for cold fixture caches. See Inspect build prerequisites. Inspect can also be built from inspect/ with ./gradlew build.

Once per clone, run sh scripts/setup-hooks.sh from the repository root to bootstrap CodeGraph and configure Worktrunk. Inspect has no separate setup.

Design and status

DESIGN.md holds the architecture and decisions: the three-grade filter, memory and learnings, runner standardization, the GitHub App split, and the validation benchmark that gates replacing a hosted review bot with Leveret.

License

AGPL-3.0-or-later.

Available Tools

5 tools
contextA

Prioritization context for reviewing a change — NOT findings: per-function cyclomatic complexity (lizard, multi-language), 12-month git churn, and last-touched date per file. High complexity in a high-churn file deserves the deepest review; use this to decide where to dig before reading code.

Inputs: repo (required), files (required).

memoryA

List the repo's review-memory entries (fingerprint, grade, reason, created, lastApplied). Use lastApplied to spot dead pricing worth deleting, and repeated same-rule entries under one subtree as candidates for promotion to a glob memory or a .leveret.yml profile rule.

Inputs: repo (required).

rememberA

Persist a graded verdict to the repo's review memory (.leveret/memory.jsonl) so the finding class never re-surfaces ungraded. Grades: priced-noise (true but the repo prices fixing it at zero) or false-positive (the claim is wrong). Only drops are stored — actionable findings are reported, not remembered. Give anchorFile + anchorLine to pin an instance verdict to its source line: the memory dies when that line changes. Omit the anchor for a class-wide verdict (fp may use a glob).

Inputs: fp (required), repo (required), grade (required), reason (required), +3 more.

scanA

Run the applicable static-analysis engines (semgrep security rules, gitleaks secrets, shellcheck, ruff, actionlint) over a change set and return normalized findings. Findings are review LEADS, not verdicts: validate each against current code. Give either base (git ref; scans base...HEAD changed files, secrets over base..HEAD commits) or an explicit files list (repo-relative). A .leveret.yml profile in the repo (or profilePath) scopes engines by path and suppresses priced rules; suppressions come back tallied with their reasons, never silently.

Inputs: repo (required), base, files, engines, +1 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.3/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool serves a distinct function: context for prioritization, scan for finding issues, ast_search for syntactic search, remember for storing verdicts, and memory for listing stored verdicts. There is no overlap or ambiguity in their purposes.

    Naming Consistency4/5

    Tool names are concise and lowercase, but ast_search uses an underscore while others are single words. This is a minor deviation, but the naming style is otherwise consistent and intuitive.

    Tool Count5/5

    With 5 tools, the server is well-scoped for a code review assistant. Each tool covers a necessary step in the workflow without redundancy, and the count is neither too sparse nor excessive.

    Completeness4/5

    The tools cover the core review lifecycle: contextual prioritization, scanning, code search, verdict memory, and memory inspection. Minor gaps exist (e.g., no explicit update/delete for memory entries), but the surface is otherwise complete and functional.

    Maintenance

    ActivityActive
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers