Fetching from the wire…
01
02
03
04
05
06
07
08
Agents2026-08-28 · source-backed
arXiv 2608.26733 presents an execution-only attack that reconstructs a hosted agent skill without ever asking the victim to reveal it, submitting crafted but ordinary tasks whose results discriminate between candidate hidden behaviors. At the weakest access level, final response plus returned files only, it recovers 86.8% of the original skill's capability across 7 skills and 4 victim models, nearly 4x the SigLeak baseline, producing an installable skill from a median of 32 calls with disclosure defenses enabled. If your business is skill-gated agent access, file secrecy and disclosure filters aren't protecting the asset.
Each link below shares sources, entities, or timing with this story.
arXiv 2608.06370 evaluated models emitting code that calls tools against JSON-schema tool calling on BFCL v4. PTC matched or exceeded the baseline in 11 of 14 models, with the GPT-5.6 family up 10.6%, and held stable under parallel execution in 13 of 14. Under context degradat...
Self-hosted agents read and write their own memory and config to function, which means an attacker can compromise one entirely through legitimate OS system calls with no exploit involved (arXiv 2607.17986). The paper builds a 23-cell attack matrix across Target, Mechanism, Gra...
Li, Huo, and Johnson show that one-way message flow between agents produces neither mimicry nor solo behavior but an entirely novel dynamical state, at identical temperature settings. It's conceptual rather than quantitative, but the implication for orchestrator-worker fan-out...
Recuris (arXiv 2608.24876) keeps a Working Memory tracking current task progress separate from an Experiential Memory of learned skills, so skill selection indexes against what the task needs now rather than the whole history. It improves 35 of 37 model-benchmark pairs, gains...
Skill-α (arXiv 2608.01678) reframes skill generation as RL over sequential edits, decomposing skill construction into individually evaluable changes. The novel signal is a rollback reward that scores each modification by comparing downstream task execution using the original s...
This one annoyed me, because I've been running the losing pattern. SWE-QA (arXiv 2608.01507) compares the sub-agent grep pattern that Claude Code, Codex and Antigravity all ship by default against a pre-built semantic index over the same repository. Semantic search answered 65...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.