How AGLedger Compares
AGLedger is accountability infrastructure. It is not a replacement for logging, observability, or governance — it is the layer underneath that ties intent to outcome.
| Approach | What it records | What it misses |
|---|---|---|
| Application logs | What happened (events, errors, traces) | What was supposed to happen, who authorized it, whether the result was accepted |
| Observability platforms | System health (metrics, traces, dashboards) | Business intent, acceptance criteria, delegation chains, verdicts |
| Governance tools | Policies, permissions, access controls | Runtime execution records, structured evidence, cryptographic proof of delivery |
| Manual audit trails | After-the-fact documentation | Real-time structure, tamper evidence, delegation traceability |
| AGLedger | What was agreed, what was delivered, whether it was accepted — signed, hash-chained, immutable | Content inspection, quality judgment — the principal decides |
AGLedger vs application logging
Logs tell you what happened. AGLedger tells you what was supposed to happen and whether the result met the original criteria. Logs are written after the fact. Mandates are locked before work begins.
A log entry says “agent called endpoint X at 14:32.” AGLedger records that a principal assigned a mandate with specific criteria, a performer submitted a receipt with structured evidence, and the principal accepted the delivery. Every step is Ed25519-signed.
AGLedger does not replace your logs. It gives them context — the mandate ID that connects every log line to the commitment it was supposed to fulfill.
AGLedger vs observability
Observability tools tell you whether your systems are healthy. AGLedger tells you whether your agents are meeting their commitments. Datadog shows you latency. AGLedger shows you that the agent was supposed to deliver a procurement report by 17:00, delivered it at 16:48, and the principal accepted it.
Use your observability stack for infrastructure health. Use AGLedger for accountability — did the agent do what it was supposed to do, and can you prove it?
AGLedger vs governance platforms
Governance defines what agents can do — permissions, policies, guardrails. AGLedger records what agents did do and whether it matched what they committed to. Governance is preventive. Accountability is evidential.
Most governance tools focus on access control and policy enforcement. They don't track whether the work that was authorized was actually completed, delivered on time, and accepted by the principal. AGLedger fills that gap.
AGLedger is not governance. It is the audit trail that proves governance worked — or didn't.
AGLedger vs manual audit trails
Manual audit trails are reconstructions. Someone writes down what happened after the fact, from memory, logs, and Slack threads. They are incomplete, inconsistent, and impossible to verify cryptographically.
AGLedger produces the audit trail as work happens. The mandate is locked before work starts. The receipt is submitted as evidence. The verdict is rendered. Every entry is Ed25519-signed and hash-chained. Tamper with any entry and the chain breaks.
The proof already exists when the auditor asks. There is nothing to reconstruct.
When to use what
“Did the service crash?” — observability
“Does the agent have permission?” — governance
“What API calls did it make?” — logging
“Did the agent do what it committed to, and can I prove it?” — AGLedger