AgentPulseCatch agent loops and drift before users do.
Paste a run log, flag loops and drift, and get a reviewable report so on-call knows what agents actually did.
No credit card required · Cancel anytime
Key Takeaways
- Loop and re-entry detection across tool calls.
- Drift / anomaly flags with a reviewable report.
- Honesty: observability decision-support — not a promise agents never fail.
Agent runs. Signals that matter.
Decision-support themes — not an uptime SLA or penetration-test certificate.
AgentPulse — definition
AgentPulse reviews pasted AI agent runs for loops, drift, and anomalous tool calls, keeps a replayable audit trail of runs, and flags issues so on-call teams can review what agents actually did.
What this is NOT
- Not a live monitor, APM replacement, or uptime certificate.
- Not a guarantee that every loop or drift in production will be found.
- Not a red-team pentest and not a statute map.
Controller / Processor — Findings name loops, drift, and odd tool calls in the paste. They are not privacy roles or a law map.
Loops burn tokens. Drift ships wrong answers.
Tool loops
Agents retry the same tool call until budgets burn.
Output drift
Answers slowly leave the intended policy or format.
Opaque traces
Logs exist, but the pasted steps are hard to read by hand.
Late discovery
Users see failures before on-call has a signal.
Paste. Flag. Review.
Decision-support review, not an uptime certificate.
Loop flags
Repeated tool-call patterns for human review.
Drift signals
Output/behaviour drift oriented hints.
Replayable audit trail
A labelled report you can inspect and export.
Honest demo vs live
Demo is labelled; live narrative needs a configured key.
Everything you need for first-pass agent run review
Built for on-call and platform teams reviewing AI agent runs.
Loop detection
Repeated tool-call patterns for human review.
Drift flags
Output/behaviour drift oriented hints.
Anomaly flags
A labelled report you can inspect and export.
Replayable audit trail
Demo is labelled; live narrative needs a configured key.
From agent run to report in 3 steps
Ingest
Capture a run trace or paste run details.
Scan
Deterministic loop/drift checks plus optional model-assisted narrative.
Report
Review a labelled report, with humans staying in the loop.
Paste. Flag. Review.
Decision-support review, not an uptime certificate.
Tool loops
Agents retry the same tool call until budgets burn.
Loop flags — Repeated tool-call patterns for human review.
Try it on this page
or generate below.
Saved history · replayable audit trail
Built for on-call and platform teams
Three checks, not customer quotes: loops, drift, and a finite run cap.
Loop detection
Agents retry the same tool call until budgets burn.
Drift signals
Answers slowly leave the intended policy or format.
Replayable audit trail
A labelled report you can inspect and export.
Honest boundaries
AgentPulse drafts loop/drift flags for pasted agent runs. Review with on-call, with no uptime guarantees.
Simple plans that scale
All paid plans include
- Agent monitoring studio
- Loop / drift checklist
- Behaviour report (decision-support)
- Email support on paid plans — no hidden add-ons for core studio use
- Limited daily AI runs (3/day · 15/mo)
- Studio demo (no extra LLM fee)
- AI included: 40 runs/day · 300/month
- No separate ChatGPT subscription required
- Priority support · 2 months free with yearly
Feature comparison
What each plan includes — quotas reflect the enforced fair-use limits.
| Dimension | Free | Pro | Enterprise |
|---|---|---|---|
| Usage | |||
| AI runs | 3/day · 15/mo | 40/day · 300/mo | 200/day · 3000/mo |
| Report export | Watermarked | Clean export | Clean export |
| Audit log | Not included | ✓ Included | ✓ Included |
| Product | |||
| Loop detection | ✓ Included | ✓ Included | ✓ Included |
| Drift flags | ✓ Included | ✓ Included | ✓ Included |
| Anomaly flags | ✓ Included | ✓ Included | ✓ Included |
| Replayable audit trail | ✓ Included | ✓ Included | ✓ Included |
| BYOK (your LLM key) | Not included | Not included | ✓ |
| SSO | Not included | Not included | Roadmap |
| Support & compliance | |||
| Support | Community | Priority | Custom SLA |
Already paid? Unlock Pro with your license
Licenses are issued manually after your payment is confirmed: we generate one per paid order and send it to the purchase email, typically within a few hours. Enter it below to activate Pro on this device — no account needed. Nothing yet? Write to lixingliangsy@163.com and reference your order email.
Your questions, answered
What does AgentPulse monitor?
Pasted AI agent runs for loops, drift, and anomalous behaviour, with a severity-scored report for review. Decision-support, not a guarantee agents never fail.
Is this legal advice or a certification?
No. Outputs are decision-support drafts. Have counsel and risk owners review before production. We do not claim ISO, CMP, or regulator certification.
Can I cancel anytime?
Yes. Self-serve plans cancel anytime; access continues until period end.
Do I need a credit card to start?
No. Start with email signup or Demo mode in the studio, then upgrade when ready.
Do I need my own ChatGPT / OpenAI subscription?
No for Free/Pro. AI runs are included in your plan (fair use) via our platform key. Enterprise can optionally bring your own OpenAI key (BYOK) — configure it at /settings (key stays server-side only).
What is Fair Use / what if I hit the AI limit?
Free is 3 runs/day and 15/month. Pro is 40/day and 300/month. Enterprise is 200/day and 3000/month. Over the cap the API returns HTTP 429. A labelled demo is opt-in, not a silent fallback.
Is checkout secure?
Payments are processed by Waffo Pancake (merchant of record).
What happens after I pay?
You receive access confirmation; fulfillment is tracked via webhook + order logs.
Quick answers
- Reviews a pasted run for loops, drift, and odd tool calls.
- Keeps a replayable audit trail (GET /api/runs).
- Produces a reviewable, exportable report.
- Push alerts near-real-time: roadmap (not shipped).
- Pricing starts at $0 (Free).
More questions
What signals?
Loops, goal re-entry, drift, and anomalous tool calls from pasted run observations.
Does it monitor agents live?
No. You paste a run and get a per-run report. This is decision-support review, not live metering.
What inputs?
Agent description plus a sample run / behavior log.
Demo vs live?
Demo is labelled; live narrative needs a configured AI key.
Who is it for?
On-call and platform teams reviewing LLM agent runs before and after shipping.
Primary sources?
Agent observability practice notes (loops, drift, tool-call audit trails).
How it compares
| Dimension | Manual | AgentPulse |
|---|---|---|
| Speed | Hours to days | Minutes per run |
| Consistency | Varies by person | Same ruleset every run |
| Output | Free-form | Structured, exportable result |
| Best for | Final sign-off | First-pass decision-support |
When not to use: Not for live monitoring. This page reads one paste.
- agent observability vs classic apm 2026 — From the AgentPulse blog.
- agent security tools compared — From the AgentPulse blog.
- Fleet launch notes — AgentPulse ships in the LX AI Micro-SaaS fleet.
People also search
Click a tab to expand it; click again to hide. All content stays on the page for search engines.
Start smarter today
Join builders using AgentPulse. Free to try — no card needed.
Trusted · Cancel anytime