On-call in 10 minutes

Your first patch PR in 10 minutes.

An on-call agent for dependency updates. It opens, tests, and merges dependency security PRs in your GitHub repos — connect GitHub, choose your policy, and your first patch PR is usually live in 10 minutes.

01Connect GitHub
02Choose your policy
03Bivouac finds an issue
04Patch PR
05Tests
06Merge

On-call in 10 minutes.

Start freeNo credit card. No sales call. No commitment.

Live walkthrough

Watch a CVE go from disclosure to merged PR.

Here is one captured run — 6 minutes, 5 stages, no engineer woken up. The same track runs on every signal we pick up.

bivouac · run · lodash/cve-2020-28500

live

09:42
1/6
CVE detected
lodash < 4.17.21 (CVE-2020-28500 in lodash.template prototype pollution)
09:43
2/6
Impact assessed
Affected dependency found in 3 repositories (catalog-graph walk)
09:44
3/6
Fix selected
Upgrade compatible with current dependency graph (4.17.21 → no peer churn)
09:45
4/6
PR created
fix: upgrade lodash to 4.17.21 (draft, awaiting tests)
09:47
5/6
Tests passed
1,248 tests passed · 12 repos in poly-repo · 0 churn
09:47
6/6
PR merged
Security issue resolved without waking an engineer
Real captured run · 09:42 → 09:47 · 0 engineer pages

Human in the loop

AI handles the routine. Engineers handle the exceptions.

Bivouac doesn't replace your engineers — it removes the repetitive dependency work that doesn't require engineering judgment. The same loop runs on every signal: investigate, fix, test, merge. Engineers stay responsible for the calls that need humans.

If it doesn't require engineering judgment, Bivouac closes it. If it does, Bivouac pages you.

The four triggers below are the only conditions that wake the on-call. This is an opt-in rule per repo and lives inside the policy pack, so it's reviewable in the same diff as the patch.

Contested decisions

When the right remediation isn't unambiguous — forward-fix vs. downgrade vs. pin — the agent opens the PR as a draft and pages a maintainer to pick.

Security-sensitive surfaces

Auth, crypto, signing, and supply-chain entry points. The agent stops and waits for human review before any patch lands in those files.

License concerns

License churn (new copyleft dependency, attribution drift, GPL contamination). The agent refuses to merge and surfaces the licensing diff to a maintainer.

Failing builds

If the test suite goes red on the patch, the agent blocks the merge, opens the PR as a draft, and writes the failure to the audit log.

Cost of one vulnerability

What does one vulnerability actually cost your team?

Same CVE. Same fix. Same end state. The gap between these two timelines is why teams buy Bivouac — not because the alert is unique, but because the path from alert to merge is.

Without Bivouac

4–6 hours

10 manual steps · engineer on the loop

With Bivouac

5–15 minutes

5 automated steps · engineer asleep

Without Bivouac

4–6 hours

10 manual steps

CVE detected
engineer receives alert
investigates
finds compatible version
creates PR
CI runs
tests fail
engineer investigates
fixes dependency conflict
PR reviewed → merged

With Bivouac

5–15 minutes

5 automated steps

CVE detected
Bivouac investigates
creates fix
runs tests
merges if green

Median across the Bivouac benchmark suite, 2026-Q2 — same CVE, same test suite, real engineer time measured end-to-end.

Built for teams that maintain their own stack

Who Bivouac is for

One product, four reasons to buy — pick the one that matches the room you're in.

CTO

Stop paying engineer-hours to maintain dependencies. Bivouac absorbs the maintenance tax so your roadmap keeps moving.

Engineering Manager

No more manually triaging Dependabot PRs at 8am. Bivouac sorts, tests, and opens the patch PR your team can merge.

Platform Team

Standardize dependency remediation across every repo. One policy, one agent, every service — without the bespoke plumbing per team.

Security Team

Turn every CVE alert into a tested, gated patch PR. Remediation lands before the alert ages out of the dashboard.

Start freeAll four start free on the same first repo. No sales call required.

How Bivouac compares

More than alerts — Bivouac closes the loop.

Dependabot and Renovate keep raising PRs. Snyk keeps flagging advisories. Bivouac is the layer that decides what to do, proves it on your tests, and ships it — only paging a human when the call is genuinely contested.

Capability	Bivouac	Dependabot	Renovate	Snyk Open Source
Monitors CVEs and breaking changes	watches CVEs and SemVer-major signals across public and private feeds	tracks public advisory feeds only	tracks public advisory feeds and SemVer drift	tracks public vulnerability feeds
Opens a patch PR	opens a draft PR against the right pin, fork, or downgrade	opens a forward-fix PR	opens a forward-fix or pin PR	no PR — surfaces an advisory
Runs your test suite	runs the project test suite before the PR is reviewable	not in scope	opt-in — automerge requires explicit maintainer config	not in scope
Merges when green	auto-merges on green with a signed audit row	no merge authority	merges only when CI passes AND maintainer config allows	not in scope
Pages a human on judgment calls	pages the on-call only on contested decisions — license, public API, security surface	not in scope	not in scope	not in scope
Handles downgrades when a forward fix is not viable	picks downgrade-as-fix when the forward upgrade is incompatible	forward-only — fails open otherwise	forward-only by default	not a remediation tool

Outcome-language summary — see pricing for what each plan actually runs in your repo.

See pricing →

For the CTO and the auditor

NIS2, CRA, SOC 2, and the dependency graph — generated as a side effect of the merge.

Every autonomous decision leaves a signed, exportable record — the same record your auditor wants to see. Compliance turns from a quarterly scramble into a query against the merge log, and the agent reasons over your full dependency graph before it picks a remediation. Bivouac produces the evidence your team hands to the auditor — it isn't itself SOC 2 / NIS2 or EU CRA certified.

NIS2 · Article 21

24/7 incident handling and supply-chain security recorded per merge — no end-of-quarter scramble.

EU CRA

Vulnerability handling, supported-product disclosure, and signed SBOMs / software-bill-of-materials per build — exportable from the merge log.

SOC 2 evidence-ready

Append-only audit trail: which signal fired, which policy pack applied, which test gate cleared the merge.

Dependency graph awareness

Reads the full transitive dependency graph before picking forward-fix, downgrade, or pin — so the chosen remediation is the one your graph actually supports.

Buyer FAQ

Five questions a buyer should be able to answer in under thirty seconds.

Plain-language answers, the way we wish someone had given them to us. Skim the questions; the answer you’re looking for is almost certainly in there. If something’s missing, ping us — we read every message.

How it works

From disclosure to merged fix — on its own shift.

Every signal runs the same three-step track. The test suite is the gate. The on-call only gets paged for contested decisions, security-sensitive surfaces, license concerns, or failing builds.

01
Watches
Continuous watch on your GitHub repositories and dependency trees.
Public and private advisory feeds, package-registry disclosures, and upstream breaking changes — scanned every five minutes.
02
Investigates
Reads the signal. Picks forward fix, downgrade, or pin against your policy pack.
SemVer, peer usage, license posture, and security-sensitive surfaces — chosen locally before any code is touched.
03
Patch, test, merge
Opens a draft PR, runs the project test suite, and auto-merges on green.
The build is the gate. Red blocks the merge. License churn, public-API breakage, and security-sensitive surfaces stop at a human — every widening is opt-in per repo.

Dimension by dimension

How Bivouac compares.

Four dimensions where the bots disagree on what counts as a fix. Bivouac’s column reflects only what runs in your repo today — anything we can’t ship on a green build, we won’t claim here.

Dimension	Dependabot	Renovate	Snyk Open Source
Detection surface	Partial
Remediation speed	Partial	Partial	No
Test-suite enforcement	No	Partial	No
Human escalation rules	No	No	No

Each cell is grounded in what each plan runs and the comparison blog series.

See the deep comparisons →

How Bivouac works

From disclosure to merged fix — on its own shift.

What happens between signal and shipped fix: four ordered steps the agent runs against every connected repo, no shortcuts, no skipped gates.

01
Detect
A CVE drops or an upstream ships a breaking change — we read it from your repo’s dependency tree, not a public feed.
02
Investigate
Trace the impact in your lockfile, your test suite, and your call sites. Pick the smallest viable fix path before any code is touched.
03
Patch and test
Open a draft PR, run the project’s test suite, and pick a downgrade when a forward fix isn’t viable. The build is the gate.
04
Merge or page
Merge on green. Page a human on contested license churn, security-sensitive surfaces, public-API breakage, or red builds.

Start free on the first repoNo credit card. No sales call. No commitment.

How Bivouac compares

Capability by capability — and where to read more.

Six properties that decide whether a dependency-watch tool actually closes the loop. Each competitor’s column links to the published comparison post so you can read the full argument behind every cell.

Capability	Bivouac	Dependabot	Renovate	Snyk Open Source
Autonomous patch PR	opens a draft PR against the right pin, fork, or downgrade	opens a forward-fix PR	opens a forward-fix or pin PR	no PR — surfaces an advisory
Breaking-change detection on dep bumps	lockfile drift and SemVer-major upstream moves are a trigger	not a trigger — advisory-driven	tracks SemVer-major upstream moves	not a trigger — advisory-driven
Fallback downgrade	picks downgrade-as-fix when the forward upgrade is incompatible	forward-only — fails open otherwise	forward-only by default	not a remediation tool
Human-page-on-judgment calls	pages the on-call only on contested decisions — license, public API, security surface	not in scope	not in scope	not in scope
Audit log / attestation	every merged patch carries a signed audit row (feed, decision, test outcome)	not in scope	not in scope	not in scope
Transitive CVE coverage	NVD + GHSA, scoped to direct and transitive per repo policy	public advisory feeds, direct-dependency scope	public advisory feeds, transitive scope	NVD + vendor-curated DB, direct + transitive scope

Each competitor column links to the in-depth comparison post — Bivouac vs Dependabot, Bivouac vs Renovate, and Bivouac vs Snyk Open Source.

Early access

Built for teams that can't afford dependency drift.

Currently onboarding early-access teams.

Bivouac is in private beta with a small set of design partners. Each team gets a named engineer, a shared roadmap slot, and a private channel into the audit log — names will appear on this page once they've signed off, not before.

Ecosystem coverage

One agent across every runtime your stack touches.

GitHub

Repos, advisories, PRs, merge guardrails.

Node.js

npm lockfiles, transitive SemVer, peer ranges.

Python

pip, pyproject, private indexes.

Java

Maven, Gradle, OSV-scored advisories.

go.mod graph, vendor trees, module proxy.

Ruby

Bundler, gemspec, RubyGems advisories.

Rust

Cargo workspace graph, crates.io advisories.

PHP

Composer, lockfile, private Packagist.

.NET

NuGet, csproj, transitive resolve.

Early-access teams

A small set of design partners is onboarding now. Names will appear here once they've signed off — we don't put logos on the page before they do.

Pricing

Per active repository. No tier zoo.

Three plans, billed monthly per active repo. All tiers ship the same audit trail and the same test-gated merge guardrails — higher plans widen what the agent handles autonomously and how fast.

Free

/ active repo / month

1 active repo · no card required

See Free on the pricing page

Starter

$19

/ active repo / month

Up to 5 repos — public feeds, forward-fix patch PRs, test-gated auto-merge.

See Starter on the pricing page

Team

$39

/ active repo / month

Up to 25 repos — private feeds, downgrade-as-fix, multi-repo policy packs.

See Team on the pricing page

Enterprise

$99

/ active repo / month

Unlimited repos — dedicated on-call routing, SBOM provenance, custom SLAs.

See Enterprise on the pricing page

Start freeFree for the first repo. No card. No call.

Volume pricing for fleets and SOC 2 / NIS2 / CRA evidence-ready audit export on every tier. See full pricing →

Talk to us

Got a fleet, a regulated environment, or a tricky policy?

Drop a few lines and someone from the Bivouac team will respond within one business day — no SDR sequence, no demo gauntlet. We'll help you size the rollout and pick the right policy pack for the repos you actually watch.

Already running Dependabot, Renovate, or Snyk? Tell us what you're paying for them — we'll show you the overlap with what Bivouac earns.

Send us a note

We read every message. Expect a real reply from a person on the team, not a sequence.

Buyer objections, answered

Five questions a buyer should be able to answer.

Pricing, test gates, what we read, the upstream registry set, and how the on-call page actually fires — the answers in plain language, the way we wish someone had given them to us.

Light the watch