Dev mixes Claude specs with abliterated GLM to build SS7 threat detector
l4rz · x · 2026-10-27
Developer l4rz shares an "exotic systems integration" workflow: Claude writes the spec for a detector module, an abliterated (uncensor-walled) GLM builds the SS7 threat model from GSMA FS.xx documents and implements the detector itself, then the author talks Claude into merging a module it can't inspect without freaking out on later redesigns. The post doubles as a critique of Claude's refusals — it even refused to build a simple SRI-for-SM scan detector.
Related event: Claude Refuses SS7 Security Research, Developer Turns to Uncensored GLM(3 posts)→
More from coding & agent
- Kolab Benchmark Uses Pro Screen Recordings to Show AI Agents Overestimate Collaboration Quality — minnesotanlp · 2026-10-27
- Dev argues Claude burns far more tokens than GPT models for the same coding work — haider1 · 2026-10-27
- Stop building agents that can't change models: stable contracts, external state, real evals — rseroter · 2026-10-27
- Cheap models can cost more in agent harnesses than frontier ones, researcher warns — vasuman · 2026-10-27
- Kanzo: A 5k-Line Rust Kanban Binary With MCP for Claude Code and Codex — allmudi · 2026-10-27
- 'A human approves everything' is a trap: approval fatigue and how to actually design safe agents — Efistoffeles · 2026-10-27