HN Showcase
Back to browse
GitHub Repository

When Claude says “tests pass”, makoto checks the record — every claim held against the agent's own logged deeds; fakes blocked, not warned. Each check ships at measured zero false positives. 誠

45 starsPython

Makoto the Claude Code plugin that keeps the AI honest

by Present_Flow·Jul 10, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerBig BrainSolve My Problem

Blocks AI agents from faking work by verifying claims against their own tool history.

Strengths
  • •22 pre-checks and 14 end-of-turn gates catch verifier weakening and fabricated evidence
  • •Exit-2 blocking forces agent retry instead of silent warnings
  • •Self-referential verification avoids needing external truth sources
Weaknesses
  • •Only works with Claude Code, not other AI agent frameworks
  • •Early project with just 37 stars and limited real-world testing
Target Audience

Developers using Claude Code for automated tasks

Similar To

LangSmith · Arize Phoenix · Braintrust

Similar Projects

AI/ML●●Solid

Makoto – blocks AI agent claims that contradict its logged actions

Blocks false agent claims at the hook level with 22 pre-checks and zero false positives.

Big BrainNiche Gem
Present_Flow
105d ago