Does PromptReports.ai have a free trial?

Yes. The free trial includes 50 credits, no card required — enough to run a full Claude Code scan and inspect detector findings before paying.

What does Token Forensics cost?

Hybrid credits + seats: Free trial (50 credits, 1 seat). Starter $99/mo (500 credits, 1 seat, daily scans + PR remediation). Team $999/mo (5,000 credits, 10 seats, signed audit bundles). Scale $5,000/mo (30,000 credits, 30 seats, SSO, hourly scans). Enterprise from $15K/mo (custom, on-prem option, BYOK signing keys).

Credits meter the metered actions — detector scans, PR remediation runs, signed audit bundle exports. Seats meter access. The hybrid model means a small team running daily scans uses few credits; an org running hourly scans on a large repo uses many.

Can I cancel anytime?

Yes. All plans cancel at any time with no cancellation fee. Access continues until the end of the current billing period.

What's a signed audit bundle?

A hash-chained turn log + HMAC-signed export bundle that your CFO and SOC2 auditor verify offline using our open-source verifier. No shared credential, no API call, no 'trust the dashboard.' SOC2-ready by construction.

Token Forensics for every team running Claude Code.

Credits meter detector runs and signed audit bundles. Seats meter access. Pick the plan that fits your monthly Claude Code spend.

16 CC-native detectors

One-click PR remediation

HMAC-signed audit bundles

MonthlyYearly

Compare All Features

Features
Usage Limits
Reports per Month
API Calls per Month
Team Seats
Storage
Report Features
AI-Powered Reports
Source Verification
PDF Export
Advanced Analytics
Bulk Generation
Customization
Custom Branding
White Label
Integration & API
API Access
Custom Integrations
Team & Security
Team Collaboration
Single Sign-On (SSO)
Audit Logs
Support
Email Support
Priority Support
Dedicated Support
SLA Guarantee

Swipe to compare plans

Frequently Asked Questions

What do credits and seats actually meter?

Credits meter detector runs (each scan against your CLAUDE.md / sessions / skills consumes credits) and signed audit-bundle exports. Seats meter access — who on your team can review findings, approve PR remediations, and download bundles. The two scale independently, so a small team running heavy nightly scans can stay on a low seat count.

How do you find waste my proxy can't see?

Helicone, Datadog, and Langfuse sit at the API boundary. They see token counts and latency. They don't see your CLAUDE.md (loaded every session), your skills (loaded but never invoked), your stop_reason patterns, or your tool-call sequencing. Our 16 detectors read that editor surface — that's where 60–80% of Claude Code waste lives.

Are the audit bundles really verifiable offline?

Yes. Every export is a hash-chained turn log + HMAC signature. Your auditor runs npx tsx verify-token-findings-export.ts <bundle> — no shared credential, no API call, no “trust the dashboard.” The verifier is open source. SOC2-ready by construction.

Can I change plans at any time?

Yes. Upgrade or downgrade any time — upgrades are prorated, downgrades take effect at the end of the current cycle. Unused credits roll over for one cycle. Enterprise contracts are annual.

What if I exceed my monthly credits?

We notify you at 80% utilization. Past 100% you can top up credits a la carte or upgrade — we never charge overage fees without your explicit consent.

Is there a free trial?

The free tier includes a starter credit pack — enough to scan a single repo and see real findings before you commit. For Team / Scale / Enterprise, contact sales for an extended trial calibrated to your monthly Claude Code spend.

What payment methods do you accept?

All major credit cards via Stripe. Enterprise customers can pay by invoice (NET-30) and procure via standard MSA + DPA.

Find the Claude Code spend your proxy can't see.

Run one terminal command. Review the findings. Ship the PR. Sign the audit bundle.