HN Showcase
Back to browse
Halftrace – See the shape of your agent's failures from your logs

Halftrace – See the shape of your agent's failures from your logs

by ruairidhwm·May 20, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerBig BrainSolve My ProblemDark Horse

Diagnoses agent failure modes from existing logs without spending new API credits.

Strengths
  • •Distinguishes between 'bimodal' and 'average' failure modes to suggest specific prompt fixes.
  • •Compares two log sets to verify if prompt changes actually improved compliance.
  • •Zero-cost analysis by parsing existing JSONL exports from OpenAI, Anthropic, or LangSmith.
Weaknesses
  • •Limited to specific JSONL export formats; custom agent frameworks may need adapters.
  • •Relies on heuristics for 'shape' classification which may miss nuanced context.
Category
AI/ML
Target Audience

AI engineers and prompt developers debugging LLM agents

Similar To

LangSmith · Arize Phoenix · Braintrust

Similar Projects

AI/ML●●Solid

A self upgrading agent that learns from failure

Agent writes its own Python tools and saves rules to avoid repeating mistakes.

WizardryBold Bet
grimm76321
102mo ago