Questions, answered.

Both — it just stops short of committing for you. Every finding ships a ready-to-apply suggested fix plus a one-click "Prompt to Fix with AI" you can hand straight to your coding agent. You never get a bare "this is wrong" — you get the change to make and the context to make it. Sigilix reviews and proposes; you (or your agent) apply and merge.

It meets your team where work already happens. Sigilix is a GitHub App for pull-request review, it triages CI failures — turning a red build into a grounded, inline root-cause comment instead of a raw log dump — and it integrates with Linear for issue triage. Findings land in the PR and issue rail you already use, not a separate dashboard you have to babysit.

No. Sigilix is a hosted GitHub App: install it, grant repo access, and reviews fire automatically on every pull request. There is nothing to add to your repo — no .github/workflows YAML, no self-hosted runner, no Docker image — and all review compute runs on Sigilix infrastructure, so your build times are untouched. You can also trigger it on demand with /sigilix review or @sigilix in any PR thread.

Same surface, different contract. CodeRabbit runs a single generalist model and posts what it says — that breadth is why it reads as noisy. Sigilix runs four domain specialists (logic, security, performance, tests) plus a synthesizer, and a finding can't post unless it earns it: every claim is checked against your code before it's shown — it has to cite specific evidence, survive a self-refutation pass, and, where it's checkable, be backed by live execution that produces a tamper-evident receipt (the VERIFIED / GROUNDED / MODEL pill). Anything it can't substantiate is demoted, not posted. You get fewer comments you don't have to re-verify — and a recall-safety floor that forbids any suppression path from dropping a real P0/P1 just to keep the review tidy. CodeRabbit tunes for breadth; we tune for findings you can act on without double-checking.

Greptile is the fair comparison — it leads with whole-repo context, and our retrieval is the same class (vector + AST + dependency graph + symbol-aware expansion). So context isn't the moat; we concede that. The moat is everything after retrieval. Greptile's pipeline is effectively two stages — the model reads context and its output is posted. Ours is five: evidence is assembled before any model speaks, the specialists must cite that evidence, and three gates — provenance, self-refutation, and live execution — run before a finding is allowed to post, each survivor carrying a proof tier you can verify rather than just trust. On top of that, a per-repo trust ledger learns what your team confirmed versus dismissed, and a provable floor guarantees no suppression path can drop a P0/P1. Greptile finds; Sigilix finds, proves it, and anchors the proof to the exact line. The numbers are on /benchmarks.

Today Sigilix runs as a hosted GitHub App on your pull requests — that's the surface. We're exploring a Cursor extension if their plugin surface supports it, but nothing is committed and there's no timeline yet. In the meantime, every finding ships a one-click prompt you can hand straight to Cursor, Claude Code, or any coding agent.

Each tier is a per-seat monthly base plus metered usage for review volume and agent tokens. There are no fixed PR windows — you pay for your team plus the work Sigilix actually does. Included usage covers most teams; beyond it, overage is billed transparently at published rates. Usage windows reset every 5 days.

Because our cost base is different. Plans built on a single first-party frontier API — a Claude Pro seat, say — burn through their allowance fast, because the underlying API is expensive. Sigilix runs on cheap, highly reliable open and frontier models — DeepSeek V4, Kimi K2.7, MiniMax M3, GLM 5.2, and more — that are industry-standard and dependable, so we can afford to give you far more headroom for the same money. You pick the model that fits your work, and your usage window resets every 5 days.

Yes, on every paid tier (Pro, Max, Ultra). Point Sigilix at Codex CLI, Claude Code, or your own SDK / API key and you pay that provider's usage directly. Your model fetches our verified context layer — the index, graph, trust ledger — so it inherits the precision without re-deriving your repo each session.

Pro already gives you the whole product — triage, PR review, the Sigilix CLI coding agent, and Deep-Research Chat over your codebase, all grounded in the context the review loop earned. Max is the same product with higher usage limits and wider access to frontier models. Ultra is Max with the largest usage windows plus org controls (SSO, audit logs, self-hosted runner).

Yes — a per-seat base plus usage. Each developer on the workspace is a seat; on top of that you pay transparent, metered usage for review volume and agent tokens (no fixed PR windows). So cost scales with both your team size and how much Sigilix actually runs.

Yes. Teams don't work in one mode, so you assign each member the tier that fits them — not everyone needs Max or Ultra — and allocate usage per seat from one workspace. A reviewer-heavy dev can sit on Ultra while the rest of the team is on Pro, all under a single bill.

Yes — switch the toggle to Annually and every paid tier drops 10%: Pro becomes $27, Max $54, and Ultra $90 per seat / mo, billed once a year. Monthly plans bill in arrears via the payment method on file; annual plans bill the seat base up front, with usage still metered transparently.