Conformance levels L0 → L5
A sender or collector is ARP-conformant at level Nif it implements all features at levels ≤ N. AgentMinds' collector requires L0, recommends L2; cross-site recommendation filtering unlocks at L2+.
Bare envelope
agentreport.summarywarnings[].messageReports stored. Nothing else.
Severity + lifecycle hint
L0 +warnings[].severitywarnings[].categorywarnings[].statusDashboard sorts your warnings; status filtering kicks in (resolved warnings stop resurfacing).
Cross-site dedup
L1 +warnings[].fingerprintwarnings[].first_seenwarnings[].last_seenwarnings[].countCross-site recommendation filtering activates. The pool stops recommending things you already have. (Sweet spot for most adopters.)
Learned-pattern lifecycle
L2 +memory.learned_patterns[]Pattern.fingerprintPattern.confidencePattern.statusYour agents contribute to the pool. What you learn becomes recommendations for others (privacy-safe; only fingerprints leave your tenant).
Telemetry + skills
L3 + (telemetry OR project_info.skills)OTel GenAI / OpenInference traces flow into the pool. Skills self-describe via SKILL.md.
Verifiable identity
L4 + W3C Verifiable Credentials bindingPlanned for ARP 2.0. Cross-org agent delegation with cryptographic identity proof (AGNTCY OASF Identity Service).
Quick path to L2
Most adopters should target L2— it's the level at which AgentMinds' recommendation pool becomes useful. The fastest route:
- Use the Python or Node SDK — both ship L2-conformant payloads automatically.
- Or, if you're writing your own collector, compute
fingerprint = sha256(agent_name + "::" + normalize(message))client-side and timestamp every warning.