Kora Doctor

Kora Doctor is an open-source CLI tool that analyzes AUDR trace files to find execution waste in AI agent runs.

KORA Doctor is a command-line utility designed to analyze execution waste in AI agent runs. It processes AUDR JSON, JSONL, or JSON array files to surface inefficiencies across tool usage, context growth, deterministic work, repeated inference, orchestration, and model choice. The tool requires no dashboard, account, or hosted service, running locally with no runtime dependencies. It prioritizes removing execution waste before suggesting model downsizing, providing scenario-based savings estimates and confidence ratings. Developers use it to narrow trace files down to specific steps for inspection. It also supports optional OpenTelemetry sidecar enrichment and privacy-preserving tool argument and result hashing.

Kora Doctor at a glance

Best for: Software engineer
Runs on: CLI, open source, self-hosted
Pricing: Free and open source under the Apache-2.0 license.
Released: October 2026
Category: AI coding tool
Website: github.com/krako-labs/kora-doctor

Key features

Analyzes AUDR JSON, JSONL, and JSON array files
Surfaces repeated tool and read calls
Tracks context amplification and token growth
Evaluates prompt cache efficiency and cache-read shares
Identifies deterministic validation candidates
Detects repeated model inference signatures
Enriches records with OpenTelemetry sidecar data
Computes scenario-based savings estimates

Use it for

Auditing AI agent traces for execution waste
Identifying redundant tool calls and repeated model inference
Analyzing prompt cache efficiency in agent runs
Detecting unnecessary orchestration overhead and context amplification

Alternatives to Kora Doctor

CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
React Doctor — Detects complex functions, repeated JSX, and production risks in React code.
Melra — Governs AI agent tool calls with local policies, approval gates, and verified receipts.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
OpenCode — Works with 75+ model providers, local models, and existing subscriptions
ODS — Automates complex local AI infrastructure setups with one command.
Kiro — Spec-driven coding with property-based tests to catch hidden bugs
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
OpenAPPA — Tracks data flows algebraically to prevent prompt injection without breaking agents.
Garak — It checks large language models for hallucinations, data leakage, and prompt injections.
arcKode — Runs fully locally via Ollama with symbol-level code indexing for compact VRAM setups.
Kolo — Teams share one browser session to watch, type, and control AI agents together.

All Kora Doctor alternatives