01
Parse
Read a local diff or GitHub PR and map the changed ranges before review starts.
Repo-First Code Review
See the repo.
See the repo.
Then see what matters.
DiffSwarm reviews a pull request like a programmer would: changed files, live issues, and plain-English activity in one local terminal app. Your code stays local. Your models stay yours.
Local-first
Review runs on your machine
BYOK
Claude Code or Codex CLI
Verified
Findings challenged before report
changed ranges
focused checks
likely issues
evidence review
evidence report
diffswarm review engine
Pipeline
diff mapped
issues tested
evidence kept
Verification
- Map changed code
- Run focused checks
- Challenge likely issues
- Attach evidence
Why It Is Different
+diff becomes focused checks
!second pass tries to disprove each issue
+report keeps only line-backed evidence
Evidence kept
Verified
Featured Workflow
A review surface built for scanning.
Preview the review workflow first, then drill into the pricing and local-first comparison when you need details.
Local Review
From changed range to verified evidence.
DiffSwarm keeps the workflow anchored to what moved, then challenges each likely issue before it reaches the final report.
Pricing
$5.99
Comparison
See the comparison
Unlimited reviews for $5.99/mo.
Bring your own Claude Code or Codex credentials. No token markup, no code proxy.
Local-first instead of per-review cloud review.
Anthropic charges up to $25 per review; DiffSwarm runs on your machine with your existing model setup.
The Flow
From Pull Request to Clear Next Steps
02
Review Files
Walk the files that moved and keep each pass anchored to real code context.
03
Verify Issues
Challenge likely findings against nearby guards, related code paths, and line evidence.
04
Ship A Report
Finish with a clean report and an interactive terminal summary.
One CLI. Two Missions.
Choose Your Vector
Bug Hunt Mode
Find Logic & Edge Cases
Track the changed files, follow the live issue list, and inspect exact code or diff context while DiffSwarm checks correctness risks.
$ diffswarm pr <url> --profile balanced
Security Audit Mode
OWASP & CWE Precision
Focus the same repo-first workflow on auth, input handling, config, and crypto changes so security issues stay readable and actionable.
$ diffswarm security <url> --min-severity high
Capabilities
Engineered for CI & Local dev
Bring Your Own Model (BYOK)
DiffSwarm supports Anthropic Claude Code and OpenAI Codex CLI. You provide the credentials to the local engine. We do not proxy your API calls, ensuring maximum privacy and zero markup on token costs.
CI/CD Integration
Run headless in actions. Use --comment to automatically post verified findings to your GitHub PR.
Auto-tuned Budgets
Pass --time-budget-minutes 60 and the reviewer will scale passes and fleet size to fit the window. Repeated expensive reruns must be explicitly approved.
100% Local Execution
The terminal app and review backend run locally. Your proprietary code never touches our servers; the only DiffSwarm network calls are entitlement checks.
Common Questions
FAQ
What API keys do I need?
You need either Anthropic Claude Code or OpenAI Codex CLI installed and authenticated. DiffSwarm uses your existing credentials to orchestrate the review. No additional API keys from us.
How much do AI tokens cost on top?
Token costs depend on PR size and your AI provider's rates. A typical balanced review uses roughly the same tokens as a normal Claude Code or Codex session. DiffSwarm never marks up or proxies your API calls — you pay your provider directly.
Does my code leave my machine?
Never. DiffSwarm runs 100% locally. The only network calls are entitlement checks to verify your subscription. Your code is reviewed on your machine by your AI engine.
How is this different from Anthropic's Code Review?
Anthropic's Code Review costs $15–25 per review, requires a Team or Enterprise plan, and runs in their cloud. DiffSwarm costs $5.99/mo for unlimited reviews, works with your existing Claude Code or Codex subscription, runs locally, and includes a dedicated OWASP security audit mode. Full comparison here.
Can I try before I subscribe?
Yes. Run diffswarm diff sample.diff --dry-run to see the full terminal UI and review flow without a subscription or LLM calls.