Specula: Automating TLA+ Formal Verification with Coding Agents to Find Hundreds of Deep Bugs
tianyin_xu · x · 2026-07-31
A research team released the beta of Specula, a tool that automatically checks system code using TLA+ based formal methods.
Specula instructs coding agents to fully automate formal specification (model + invariants) and mode-code conformance, overcoming major practical barriers to adopting formal methods. The researchers found Specula highly effective at uncovering deep bugs that require formal reasoning, having already identified hundreds of them. The tool operates in a push-button automated fashion.
More from coding & agent
- Open-Source MCP Server Manages Multiple SSH Connections for Claude Code — tom_doerr · 2026-07-31
- The Hidden Cost of Multi-Agent Orchestration: Context Pollution — rseroter · 2026-07-31
- Testing GitHub Copilot: Generating Robot Inverse Kinematics Code from Scratch — kamathsblog · 2026-07-31
- LangChain Founder Highlights OpenWiki as Long-Term Memory for Codebases — BraceSproul · 2026-07-31
- How AI Agents Learn from Failure: Inside the CodeClash Auto-Research Loop — agihouse_org · 2026-07-31
- Can AI Agents Write NeurIPS Papers? CRUX Benchmark Sparks Adversarial Collaboration — sethlazar · 2026-07-31