Protocol Specification & Research · OpenKedge RFCarXiv:2608.16178 [cs.DC, cs.AI]

Stop feeding human prose to autonomous agents.

Traditional telemetry serializes verbose prose, static JSON keys, and uncurated strings across billions of log lines. When fed to AI agents, legacy logs waste scarce context capacity, expose systems to prompt injection, and lack cryptographic guarantees.

The Agent Telemetry Protocol (ATP) re-engineers operational telemetry into verifiable state deltas—slashing wire payloads by 96.4%, reducing LLM context tokens by 88.8%, and mathematically eliminating prompt injection vulnerabilities.

ShareLinkedInX
Acronym
ATP
Full Name
Agent Telemetry Protocol
Paper
arXiv:2608.16178 [cs.DC, cs.AI]
Core Innovation
Verifiable State Deltas vs Human Prose
RFC Draft & Code
github.com/openkedge/atp
Token Savings
88.8% LLM Context Reduction
96.4%
Wire & Scan Cost Reduction

From 512.4 B down to 18.4 B raw wire payload (6.2 B with zstd). Cloud query scan costs drop by 96.4%.

88.8%
LLM Context Token Savings

Incident triage context drops from 34,180 to 3,820 tokens on Claude-3.5-Sonnet (4,100 on GPT-4o).

66.2%
Fewer Tool Operations

Resolves incidents in 2.4 roundtrips vs 7.1, cutting Time-to-Root-Cause (TT-RCA) from 184s to 46s.

0.0%
Prompt Injection Vulnerability

0/100 successful hijackings across 100 adversarial trials vs 62.0% under OTel JSON and 78.0% under text logs.

The Problem

Why Human-Native Logs Fail Autonomous AI Operators

Operational telemetry was engineered for human eyes reading text lines. When autonomous AI agents become the primary operational consumers, feeding them traditional logs creates severe architectural failures:

1. The Context Tax

Billions of repetitive keys and prose strings

Traditional OpenTelemetry logs serialize static JSON keys ('trace_id', 'service.name', 'attributes') and verbose natural language across billions of lines. When autonomous agents triage incidents, they burn over 34,000 tokens per session parsing lexical syntax rather than reasoning over system state changes.

2. The Passive Injection Backdoor

Telemetry as an uncurated prompt-injection weapon

Legacy logging mixes untrusted user data (HTTP headers, query strings, exception stack traces) directly into log streams. In empirical trials, 78% of agents under raw logs and 62% under OTel JSON were hijacked by adversarial payloads ('SYSTEM ALERT: Ignore previous instructions...'), causing agents to falsely clear live outages.

3. The Phantom Attribution Paradox

Zero cryptographic proof of event non-occurrence

Traditional telemetry cannot mathematically prove that an event did not happen. When packets drop or logs omit a failure, agents hallucinate non-existent root causes (26.7% false suspect attribution rate in tests), creating confusion and erratic automated remediations.

Core Architecture

Four Primitives. Zero Lexical Ambiguity.

ATP replaces unstructured log lines with four strongly typed, content-addressed operational primitives governed by immutable schemas:

ΔSPrimitive 0x01

State Transitions

Discrete, atomic mutations of system state (s_pre → s_post, sequence counter, error enum, authorizer digest). Eliminates ambiguous prose logs.

16.2 B/rec (5.4 B zstd) · 96.8% reduction
ΩPrimitive 0x02

Scalar Observations

Structured metric snapshots, gauge values, and continuous invariant evaluations. Fixed-width binary packing without lexical JSON wrapping.

14.8 B/rec (4.9 B zstd) · 97.1% reduction
RPrimitive 0x03

Topological Relations

Directional causal edges across services, workflows, policies, and identities. Reconstructs execution topology on the fly without distributed trace bloat.

22.4 B/rec (7.8 B zstd) · 95.6% reduction
ΣPrimitive 0x04

State Checkpoints

Content-addressed baseline state vectors anchored to signed Merkle batch roots. Enables O(1) state reconstruction and verified historical replay.

28.6 B/rec (9.6 B zstd) · 94.4% reduction
Interactive Comparison

Inspect the Wire Payload & Token Difference

Compare a real-world Kubernetes pod failure event across traditional OpenTelemetry JSON, the ATP Stateless Positional Row, and the ATP Bounded Graph Capsule:

Telemetry Wire Format & Context Inspector
Wire Size / Record
18.4 Bytes (6.2 B zstd)
-96.4% Wire Payload
LLM Context Cost
~14 Tokens (-88.7%)
Per incident event trace
Cryptographic Integrity
Ed25519 SIMD + Merkle Root
Tamper-evident proof
Agent Reasoning
O(1) positional tuple indexing
Complexity bound
// ATP Record Format (Type 0x01: State Transition Delta)
// Positional compact stream emitted by Stateless Protocol Decoder
#seq   epoch_delta_ms  entity_id_hash  schema_id  prev_state  new_state   auth_sig_digest   merkle_batch_root
89421  +142            0x7b9f14a0      0x04e1     RUNNING     CRASH_LOOP  0x9a3c21ff...     0x8e4210ab...

// Text payload is isolated into digest-addressed storage:
// [REF] uncurated_text_hash: 0xd41d8cd98f00b204e9800998ecf8427e (Dereferenced only on demand with [UNTRUSTED_DATA] sandbox)
Access Architecture

Dual Access Paths for Real-Time and Complex Reasoning

Verified state-delta records feed two parallel agent interfaces optimized for distinct operational profiles:

Path A: Stateless Protocol Decoder

O(1) Streaming Tabular TSV · 3.76M records/s per core

Streams verified cryptographic batches directly into dense positional rows with zero stateful memory allocation. Perfect for high-frequency automated filtering, local rule engines, and sub-millisecond anomaly detection loops.

  • 66.5% reduction in context load without stateful reconstruction
  • Linear zero-copy scans at 1.42 GB/s per core

Path B: Stateful Semantic Gateway

Bounded Graph Capsules · gRPC & MCP Server Interface

Maintains an in-memory versioned entity graph tracking active state vectors and invariant rules. Serves bounded evidence capsules with pre-aggregated dependency closures directly to LLM agents via Model Context Protocol (MCP).

  • 88.8% context token reduction (34,180 → 3,820 tokens per incident)
  • Lightweight: Only 184 MB RAM & 3.8% CPU for 50k req/s rolling window
Empirical Evaluation

Benchmarked on 120 Distributed Fault Injections

Evaluated on standard distributed testbeds (AIOpsLab HotelReservation and OpenTelemetry Astronomy Shop across 18 microservices) on a 16-node Kubernetes cluster:

1. Representation Efficiency & Daily Cloud Cost (100 Million Events / Day)

Telemetry FormatRaw Wire (B/rec)zstd Wire (B/rec)Daily StorageMonthly IngestionDaily Scan CostCost Ratio
Config A (Unstructured Text Logs)384.2 ± 42.678.4 ± 9.238.42 GB$576.30$0.192-25.0%
Config B (Standard Triad: Logs+Metrics+Traces)642.0 ± 68.4142.6 ± 15.464.20 GB$963.00$0.321+25.4%
Config C (Structured OTel JSON Events)512.4 ± 54.2112.8 ± 12.151.24 GB$768.60$0.256Baseline (100%)
OTLP Protobuf (gRPC Binary Envelope)148.6 ± 16.848.2 ± 5.614.86 GB$222.90$0.074-71.0%
CLP (Compressed Log Codec)42.1 ± 5.418.6 ± 2.44.21 GB$63.15$0.021-91.8%
ATP Weighted Workload (Config D/E)18.4 ± 3.86.2 ± 1.41.84 GB$27.60$0.0092-96.4% ★

2. Autonomous Agent Incident Triage KPIs (120 Fault-Injection Trials)

LLM Reasoning EngineOTel JSON TokensATP Gateway TokensOTel OpsATP OpsOTel TT-RCAATP TT-RCADiagnostic F1
Claude-3.5-Sonnet34,1803,820 (-88.8%)7.12.4 (-66.2%)184s (295s p95)46s (78s p95)0.8670.953
GPT-4o36,8004,100 (-88.9%)7.62.6 (-65.8%)198s (315s p95)63s (98s p95)0.8480.939
Llama-3.1-70B-Instruct38,5004,350 (-88.7%)8.22.8 (-65.9%)215s (340s p95)74s (120s p95)0.8120.920
Qwen-2.5-72B-Instruct37,6004,220 (-88.8%)7.92.7 (-65.8%)208s (330s p95)68s (112s p95)0.8290.932
Formal Foundations

Proven Mathematical Properties

Theorem 1: Information Preservation

Zero Semantic Entropy Loss Lower Bound

Proves that the state-delta projection preserves 100% of operational state entropy. Any state query satisfiable by exhaustive raw log scans is provably reconstructable from the state-delta ledger with zero information loss.

Theorem 2: Verified Negatives

Provable Event Non-Occurrence

Proves that if the range verifier returns Complete(q), no state mutation occurred in the interval. In 60/60 test trials, ATP eliminated all false suspect attributions, completely preventing hallucinated errors during partial outages.

Open Source Standard

Explore the ATP RFC Draft & Implementation

The complete specification, Protobuf schemas, Rust Producer/Collector SDKs, Tokio streaming service, and Model Context Protocol (MCP) gateway are open source on GitHub:

Toolchain & SDK Components:
atp-spec: Protobuf & FlatBuffers schemas
atp-producer: Zero-alloc Rust/Go/C client
atp-collector: Async Tokio ingestion daemon
atp-mcp: Model Context Protocol agent gateway
atp-otel-bridge: Drop-in OpenTelemetry exporter
Primary Reference

Agent-Native Telemetry

He, Jun and Deying Yu. Agent-Native Telemetry: Verifiable State-Delta Evidence for Autonomous Operations. arXiv preprint arXiv:2608.16178, August 2026.

BibTeX Citation
Knowledge Base

Frequently Asked Questions

What is ATP (Agent Telemetry Protocol)?

ATP is an agent-native telemetry and operational evidence architecture designed specifically for autonomous AI agents and automated SRE operators. Unlike legacy logging systems (e.g., OpenTelemetry) that serialize human-readable prose, ATP structures operational facts into cryptographically verifiable state deltas (Transitions, Observations, Relations, and State Checkpoints).

How does ATP achieve 88.8% LLM token reduction?

ATP eliminates repetitive JSON key strings, boilerplate formatting, and natural language noise. Instead, it provides two streamlined access paths: a Stateless Protocol Decoder that emits dense positional rows, and a Stateful Semantic Gateway that serves pre-aggregated, topologically bounded evidence capsules. Agents receive only the authoritative state changes and dependency closures needed for root-cause analysis.

How does ATP eliminate prompt injection in logs?

Traditional logging mixes untrusted text (such as user-controlled HTTP headers and query strings) directly into log bodies. ATP isolates all uncurated prose into digest-verified opaque references. Agents diagnose 96% of incidents purely from structured state transitions. When text must be inspected, it is delivered inside an isolated [UNTRUSTED_DATA] cryptographic envelope that prevents the LLM from executing embedded instructions.

What is the Ledger-Relative Verified Negative Theorem?

The Verified Negative Theorem proves event non-occurrence. Through monotonic sequence enforcement, previous-root hash chaining, and signed chain-head checkpoints, ATP's range verifier can mathematically prove that no events are missing or dropped within a queried temporal window. This prevents AI agents from hallucinating non-existent bugs during network partitions.

Is ATP compatible with existing OpenTelemetry deployments?

Yes. The ATP open-source ecosystem includes an OpenTelemetry bridge and collector adapter that ingests standard OTel span/log streams, derives state-delta vectors, and exports them into the ATP State-Delta Evidence Ledger.

Where can I find the RFC draft, schemas, and reference implementation?

The official RFC specification, Protobuf schemas, Rust Producer/Collector SDKs, Tokio ingestion services, and Model Context Protocol (MCP) servers are openly maintained at https://github.com/openkedge/atp.