CL CodeAgent Ledger

AI Coding Agent PR Liability & Evidence Log

AI coding agent audit log for PR evidence before risky code ships.

Connect GitHub PRs and agent run logs to see file-level attribution, test evidence, approval ownership, and risk summaries in one merge-ready record.

View pricing plans

Default checkout: Team annual, $1,194/year. Annual billing is 50% lower than monthly.

4 actors human, agent, sub-agent, auto-fix 6 risk classes auth, payment, PII, infra, secrets, legal 1 export incident and compliance pack

PR #8421

Agent evidence preview

High-risk diff
src/auth/session.ts Agent Needs sign-off
api/billing/webhook.ts Sub-agent CI linked
lib/pii/redaction.ts Auto-fix Review gap
cloudflare/wrangler.toml Human Approved
CIpassed
Lintpassed
Review1 gap
Manualrequired
  1. AuthorizedMaintainer approved agent run
  2. ChangedAgent edited 9 files, 3 sensitive
  3. EvidenceCI, lint, review, manual sign-off bound
AttributionHuman, agent, sub-agent, and auto-fix file ownership.
Evidence bindingCI, lint, review, manual sign-off, and failed-test overrides.
Sensitive flagsAuth, payment, PII, infra, secrets, and legal-page changes.
Incident exportRollback steps, impacted files, user-visible changes, and Q&A.

PR evidence board

Make the risky part of AI-generated code visible before merge.

CodeAgent Ledger reads the pull request diff and the agent run record, then builds a single audit surface for reviewers, security owners, vendors, and customers.

  • AI attribution separates human edits, agent edits, sub-agent work, and automated repair commits.
  • Risk files are marked red when they touch auth, payment, PII, infrastructure, secrets, or legal pages.
  • Every high-risk file can require CI, lint, code review, and manual sign-off evidence before merge.
  • Exports keep the useful record without leaking raw prompts, secrets, or unrelated internal context.
See review evidence workflow
CodeAgent Ledger audit evidence dashboard showing PR diff risk, attribution, timeline, and test proof.
Audit timeline, attribution, test proof, and merge risk in one PR-centered view.

Operating model

From agent run to defensible merge record.

The workflow stays close to GitHub so engineers can review evidence without leaving the PR context.

1

Connect

Link GitHub PRs and the agent run logs your team already uses.

2

Classify

Group file changes by human, agent, sub-agent, and automated repair ownership.

3

Bind evidence

Attach CI, lint, security review, code review, and manual sign-off to risky files.

4

Approve

Record who authorized the agent, who merged, and who overrode any failed test.

5

Export

Create incident packs, customer Q&A, rollback notes, and compliance reports.

Pricing

Default Team annual checkout is ready.

Annual billing is selected by default and gives a 50% savings versus monthly billing.

Team annual: $1,194/year for 50 repos and audit exports.

Solo

For a founder or senior engineer who needs AI change evidence on a few active repos.

$24.50/mo

Billed annually as $294. Monthly price before annual discount is $49.

5 repos

  • AI attribution ledger
  • Risk file flags
  • CI and lint evidence links
  • Monthly incident pack export
Default

Team

For engineering teams and outsourcing vendors shipping agent-assisted PRs every week.

$99.50/mo

Billed annually as $1,194. Monthly price before annual discount is $199.

50 repos

  • Everything in Solo
  • Audit exports
  • Reviewer and override chain
  • Vendor and client questionnaire builder

Org

For platform, security, and compliance teams standardizing controls across many repos.

$299.50/mo

Billed annually as $3,594. Monthly price before annual discount is $599.

300 repos + API

  • Everything in Team
  • API and queue jobs
  • Custom sensitive-change policy
  • Priority support and evidence retention controls

Questions

Common buyer questions.

Does CodeAgent Ledger replace human code review?

No. It makes human review more accountable by showing attribution, risky files, missing evidence, approvals, and failed-test overrides before merge.

Can this help software outsourcing companies?

Yes. Vendors can export client-safe AI coding evidence, explain AI usage boundaries, and answer customer questionnaires with a consistent record.

Does checkout keep this page open?

Yes. NOWPayments opens in a centered payment window while the original page remains open behind a blurred payment state.

Which plan is selected by default?

Team annual is the default. It covers 50 repos, audit exports, approval chains, and questionnaire support.