Skip to main content

AI DevOps · India

Retrace

The execution replay engine for AI agents

Visit retraceai.tech (opens in a new tab)0 votes. Log in to vote for Retrace in print
  • Community votes0
  • DR by Ahrefs5 +2 (up) since last check

Ahrefs Domain Rating set in type 17 Oct 2026, reset 15 Oct 2026.

Retrace metrics

Set in type for retraceai.tech: Ahrefs Domain Rating and Majestic link data, with the change since the last edition.
MetricValue
Domain Rating5 +2 (up) since last check
Trust Flow20 no change since last check
Citation Flow25 no change since last check
Referring domains115 no change since last check
External backlinks367 no change since last check

Ahrefs Domain Rating set in type 17 Oct 2026, reset 15 Oct 2026. Lead topic: Business (20). Majestic set 9 Oct 2026, reset 9 Oct 2026. Link data typeset by Majestic. Trust Flow, Citation Flow and Topical Trust Flow are trademarks of Majestic-12 Ltd.

Retrace domain colophon

From the registry's public RDAP record, checked 16 Oct 2026, 14:20 UTC.
Domain age4 months
Registered
Expires
RegistrarNamify Domains Inc
DNSSECNot signed

Full colophon for retraceai.tech

Record every LLM call and tool invocation your AI agent makes, replay executions step-by-step, fork from failures, and prove fixes work before shipping.

The story so far

Screenshot of the Retrace homepage
Homepage of Retrace

Retrace is an execution replay engine for AI agents that records every model call, tool call, decision, and error as an interactive timeline. Developers can replay failures step-by-step, fork from the exact step that broke, and cascade-replay downstream execution to debug the real cause instead of the symptom.

The platform provides a full reliability loop: record agent executions with one decorator, replay to understand what happened, fork and modify inputs to test fixes, and prove changes actually improved behavior before shipping. It works with any Python or TypeScript agent framework including LangChain, CrewAI, and LlamaIndex, and automatically instruments calls to OpenAI, Anthropic, and Google Gemini.

Beyond debugging, Retrace includes runtime enforcement with guardrails that halt agents when they exceed budget or loop limits, detection of hallucinations and drift, evaluation gates that fail builds when behavior regresses, and support for multi-agent systems with session tracking and agent topology graphs.

Typeset features

  • Record every LLM call, tool call, and error with one decorator
  • Step-by-step replay of agent executions
  • Fork and cascade-replay from any failing step
  • Prove-the-fix verification before shipping
  • Runtime guardrails and circuit breakers
  • CI/CD regression gates
  • Multi-agent session tracking and topology graphs
  • Automatic hallucination and drift detection
  • Semantic search across agent runs
  • PII auto-redaction on every plan

Pricing

Pricing model: Freemium

Free plan includes 1,000 traces/mo and 7-day retention. Starter at $29/mo adds 10,000 traces/mo and 30-day retention. Pro at $99/mo includes 50,000 traces/mo, 90-day retention, and unlimited fork replays. Teams at $399/mo provides 500,000 traces/mo, 365-day retention, up to 10 users, and 1,000 prove

See Retrace's full pricing (opens in a new tab)

Compare and explore

Questions and answers

How long does setup take?

Under 2 minutes. Install the SDK, add one decorator, and traces stream immediately with no infrastructure to manage.

What languages and providers are supported?

Python and TypeScript SDKs with auto-instrumentation for OpenAI, Anthropic, and Google Gemini. Works with any agent framework: LangChain, CrewAI, Vercel AI SDK, AutoGen, and LlamaIndex.

How does fork & replay work?

Select any span in a trace, modify its input, and Retrace cascade-replays from that point forward. Context from the fork flows into subsequent LLM calls, and you get a side-by-side diff with cost and latency deltas.

What are guardrails?

Runtime policies that monitor your agent in real-time. Set cost budgets, loop detection, context overflow limits, or latency caps. When a limit is crossed, the agent receives a HALT command, stopping runaway loops or budget blow-outs at the limit.

Is my data secure?

Data is encrypted in transit with TLS and at rest with AES-256-GCM. API keys are SHA-256 hashed. PII auto-redaction runs on every plan. Tenant isolation is enforced at the application layer with all queries scoped per user.

Find Retrace online

Leave a review for Retrace

No review of Retrace set in type yet. Set the first.

Your rating

Reviews are free to set. None are paid for. Read the proofs

Readers' notes · Reprints · Share the page · Your company in print? Claim it · Report a misprint

More on this profile

Retrace has a free listing. The owner can unlock the full profile with a paid launch for $5:

  • Founder story (locked)
  • Milestones (locked)
  • Demo video (locked)
  • Dofollow website link (locked)

More pages like this

  1. Number 1: OpenLIT logoOpenLIT

    Open-source agent harness engineering platform built on OpenTelemetry

    AI DevOpsIndia

    DR 65 no change since last check 0 votes. Log in to vote for OpenLIT in print

  2. Number 2: NeKu AI

    Enterprise AI platform for SAP processes with ready-to-run workflows and custom multi-agent systems

    AI DevOpsTurkey

    DR 5 −1 (down) since last check 0 votes. Log in to vote for NeKu AI in print

  3. Number 3: AnchorShell Relay logoAnchorShell Relay

    Smart AI routing gateway for LLM traffic management

    AI DevOps

    DR 0 no change since last check 0 votes. Log in to vote for AnchorShell Relay in print
Help centre