Specula: open-source agentic tool finds deep concurrency bugs via auto-written TLA+ specs
tianyin_xu · x · 2026-09-23
Alongside TLAPS-Bench, the author promotes the team's agentic tool Specula (specula-org, 468 stars).
- Specula uses coding agents to automatically write TLA+ specifications of a target system, including invariants capturing correctness properties and formal models of the implementation
- It then model-checks the specs and reproduces violations at the code level, targeting deep bugs in concurrent and distributed systems
- It has found deep bugs in multiple open-source projects, with a maintained bug list and a case-study archive including models and reproduction evidence
- The tool now supports a CI feature and is under active development
More from coding & agent
- Peter Yang's Opus 5.5 hands-on: recreating Disney's Soaring flight and calling Claude's comeback — petergyang · 2026-09-23
- Dev: if your eval costs $20 for 100 runs, the problem set isn't hard enough — pvncher · 2026-09-23
- He used AI to build his 2019 game idea — it worked, but the idea wasn't as good as he thought — BLUECOW009 · 2026-09-23
- Open-source Agent Message Board gives parallel coding agents a shared SQLite forum — airesearch12 · 2026-09-23
- A good eval has a shelf life of maybe 3 months, practitioner argues — pvncher · 2026-09-23
- Drowning in markdown: developers seek a SCRUM-like process for agentic coding — nezvanovova · 2026-09-23