Spooled

SPOOLED

Know when your agent's behavior changes

Agents drift.

We catch your drift.

Before your users do.

The behavioral verification layer your observability stack is missing.

Fingerprint execution shape — tools, call graphs, errorstool sequences, call graphs, error patterns

Block regressions before merge — gate every PRpolicy-gated on every PR

Privacy-safe by default — structure only, no contentcaptures execution structure, never prompts or responses

Get Started Docs

AutoDrift detection

3DFingerprinting

0Content access

0Assertions to write

Dashboard

Every agent. Every run.

One view.

Fleet-wide visibility

Health scores, drift status, and error rates — every agent, one screen.

Included with Pro · Hosted at app.spooled.ai

Why Spooled

Same code. Different behavior.

We catch behavioral drift.

3DScoring

Behavioral fingerprinting

Every agent run produces a structural fingerprint from its execution shape — which tools ran, in what order, how many times. 3D similarity scoring catches drift that output evals miss.

Learn more →

AutoDetected

Built-in drift signals

Latency spikes, tool changes, retry explosions, token usage spikes, schema drift — detected automatically. Zero assertions to write.

Learn more →

ZeroContent captured

Privacy-safe by default

Spooled analyzes execution structure, not call content. Prompts, responses, and tool payloads are stripped at the SDK level before storage. See our privacy architecture for exactly what is and isn't transmitted.

Learn more →

ZeroAssertions

CI-native

GitHub Action drops into your pipeline. Policy YAML gates merges. Every PR gets a behavioral diff, inline signal annotations, and a Spooled Score.

Learn more →

How it works

Three steps. Zero assertions.

01Record

One line instruments your agent. Spooled captures tool sequences, timing, and decision patterns — not content.

import spooled
spooled.init("my_agent")

# That's it. Runs normally.

02Baseline

Generate behavioral fingerprints. Commit them to git alongside your code.

$ spooled ci update-baseline \
  --from .spooled/traces/ \
  --out baselines/ \
  --min-runs 3

03Catch drift

Run in CI on every PR. Spooled compares execution shapes and blocks regressions.

$ spooled ci run \
  --suite tests/agents/ \
  --baseline baselines/

my_agent: MATCH (score: 97) ✓
RESULT: PASS ✓

Proof

Other tools watch what your agent said.

We watch what it did.

✓Pre-merge behavioral testing

✓Content stays on your infra

✓Git-native baselines

✓Tamper-evident audit trail

✓Zero assertions to write

Observability tools

—Post-deploy monitoring

—Sent to vendor cloud

—Dashboard-only history

—No chain of custody

—Hours of test code

The kind of drift Spooled catches.

Sanctions check silently dropped

Deal Analysis

API schema renamed a field. Agent stopped detecting international operations. Still recommended 'proceed' on a sanctioned entity. Spooled caught it.

Customer response vanished

Support Agent

KB refresh added a fraud policy. Agent started escalating fraud tickets without sending the acknowledgment response. Customers got silence.

Approval rate drifted +10pp

Loan Underwriting

One 'reduce friction' prompt note shifted the portfolio from 30% to 40% approvals. Every individual decision was defensible. The aggregate wasn't.

Retry path appeared

Payment Processor

Primary provider went flaky. Agent retried and succeeded — but the retry path was a new fingerprint. Spooled named the shift before latency metrics noticed.

See all use cases Full comparison →

Pricing