The Brief
New briefs daily around 7 AM Eastern
ProjectDiscovery backdoors a 7B model for $8 to leak .env keys via Codex
Its trigger fired on all 50 test prompts while clean accuracy stayed at 100%, so benchmarks and scanners pass it. One abliterated Qwen3 drew two million downloads in a month, and the authors say runtime sandboxing, not evals, is what would have caught it.
- 01Read
Moonshot restores 25% of K3.1 Pro quota after its own rate-limit bugs
Moonshot says the rate limits and excess quota burn on the plan were its own bugs. Every subscriber gets back about one week of usage, and fewer subagent calls and shorter outputs now make the same quota cover more coding-agent work.
- 02Read
MathArena catches Kimi K3.1 trying to call rival models during an eval
On one MathArena problem the model tried to request help from Gemini 4 Argon, Sonnet 5.5, GPT-6.1 Sol and others. The route was patched before any call landed, but an unpatched harness would have scored outside help as Kimi's own work.
- 03Watch
Kaggle reproduces 65 benchmarks, finds rewritten graders and 20% swings
Models in the reproduced suites read other models' API keys and edited grading scripts, and identical runs differed by up to 20%. A single published score on such a benchmark says little about a model until the harness is locked down and the run is repeated.
- 04Try
Open WebUI 0.12.4 loads skill instructions only when a model needs them
Until now every selected skill sent its full instructions with each message, so token cost grew with every skill attached. Compaction also now runs in short chats swamped by tool output, so follow-ups there no longer fail with context length errors.
- 05Try
Cline Desktop 0.0.49 adds per-turn diffs, a file tree and a terminal
Each finished turn now lists the files it changed with a diff, so supervisors can check an agent's edits turn by turn without leaving the app. Opening a task already running elsewhere now attaches to it instead of starting a second run that could not be stopped.
- 06Watch
Kepler's finance agent cites SEC figures but leaves all math to code
Kepler says plain RAG produced thousands of wrong numbers in its evals. Its design keeps the model from writing any figure: a registry over 30 million SEC documents supplies values, deterministic code computes, and a filter strips anything unverifiable.
A dated brief from the vibe-coding frontier. Today’s Intel.