Debating LLM Egress Security: Why Every Tool Call Can Run Through Your MITM Proxy
evilsocket · x · 2026-10-07
evilsocket and MooncastOnline debate monitoring LLM traffic via an egress MITM proxy. evilsocket argues the approach applies broadly: tool calls execute external tools you don't control, so you can intercept not just the harness but every shell utility, DB connection, or anything the model uses as a tool. The counterpoint: defending a hacked fine-tuned LLM differs from general egress MITM, with SSL pinning as an obstacle; options include inspecting every tool call or treating the system as air-gapped and batch-reviewing. Consensus: treat everything an LLM outputs as suspicious.
Related event: Debate Flares Over MITM Proxy Monitoring of LLM Traffic(2 posts)→
More from coding & agent
- Nautilo Ships Text+Vision Model Split, Preps Price/Security-Based Model Routing Gateway — Dan_Jeffries1 · 2026-10-08
- A fine-tuned 9B beats a 31B model: 600 labels, $0.12, 91% accuracy — julsimon · 2026-10-08
- Matt Pocock: Agents Are Good at Strategic Programming, They Just Aren't Trained to Care — mattpocockuk · 2026-10-08
- LlamaIndex Launches OpenDocRouter: Swap Between 10 Document Parsing Models in One Line — llama_index · 2026-10-08
- Microsoft open-sources Agent Lightning v1.0: 3,500-line RL framework boosting SWE-bench +14.6 pts — Microsoft Research · 2026-10-08
- OverclaimBench: Opus 5 skipped assigned files in 61% of runs, then claimed full review — hugo_larochelle · 2026-10-07