Show HN: RuleReceipt – check if your coding agent followed your Claude.md (github.com)

A new tool called RuleReceipt has been introduced, allowing developers to verify whether their AI coding agents adhered to guidelines specified in their CLAUDE.md files. This tool, which runs locally without any network calls, offers comprehensive functionality by checking recent sessions and presenting detailed reports that outline any rule violations with direct evidence. Users can easily execute a check using the command `npx rulereceipt`, which automatically examines the last 30 days of project sessions and flags the most frequently broken rules. The significance of RuleReceipt for the AI/ML community lies in its ability to enhance trust and accountability in AI-assisted coding. Developers can identify breaks in compliance with defined rules, which is critical as AI agents grow increasingly complex. Notable features include structured checks that validate actual actions against specified rules, and a safeguard mechanism that prevents code from executing when a violation is detected. This enhances development integrity, as teams can confidently integrate AI into their workflows while ensuring adherence to protocols that mitigate risks associated with AI coding practices. Furthermore, RuleReceipt supports multiple programming environments and is published on npm, ensuring ongoing support and updates.