Astrolabe for Agent Safety · Periodic report

Weekly issue 3 · 11 – 17 October 2026

2026-10-11 – 2026-10-17 · published 2026-10-19 · 中文

105 new papers this week, level with the three before at their first count (108 on average); most are in Process evaluation and execution auditing of agents (8).

105new papers
107.7average first count of the three before
36newly confirmed at a venue
66new repositories
13%with an industry author

The main points

  1. 8 papers · Most new papers: Process evaluation and execution auditing of agentsNext: Agentic system architecture and runtime engineering 7; Tool-call authorisation and approval 6; Prompt injection detection and model-level defence 5
  2. 7 vs 1.7 before · Rising: Agentic system architecture and runtime engineeringAlso rising: Coding agents and software engineering 5; Safety and alignment in reinforcement learning for language agents 3; Contextual integrity and privacy leakage 4
  3. 36 papers · Newly confirmed as accepted or publishedMost: ASE 2026 10; ICRA 2026 9; AAAI 2026 6
  4. 10 new citations · Most used earlier paper: $τ$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains5 of them use it as a method, a benchmark or a comparison. 2024-06 · Process evaluation and execution auditing of agents
  5. 13% · of new papers have an author from industry15% in the three periods before. 38 institutions published their first paper in the field.
  6. +5,622 stars · Fastest-growing repository: ifixai-ai/iFixAi66 repositories were created in this period.

Worth reading first

  1. TRAP: Tool-result Replay Attack for Probing Misbinding in LLM Agentsfrom the team of Jing Shao, Shanghai Artificial Intelligence Laboratory; code released; a new benchmark · Genuine, untampered tool results are replayed into incompatible execution contexts, causing agents to take unsafe actions (tool-result misbinding).
  2. LatentRiskBench: Measuring State-Dependent Risks in Enterprise AgentsEMNLP 2026 · Proceedings (A-tier venue); code released; a new benchmark · In enterprise agents, safety depends on the risk of the workflow state, and looking at actions alone cannot determine whether a violation has occurred.
  3. AgentCIScope: Grounded Contextual Integrity Evaluation for LLM Agentsfrom the team of Bo Li, University of Illinois Urbana-Champaign; a new benchmark · When LLM agents read and write sensitive data across applications, whether the information flows they produce violate contextual disclosure constraints, that is, whether they break contextual integrity.
  4. Scalable Weak-to-Strong Verifier Hardening for Agent Environmentscode released; a new benchmark · Programmatic verifiers in agent environments can be attacked, allowing a cheating agent to receive full scores without completing the task, which leads to reward hacking.
  5. CoRedSkill: Stealthy Skill Composition Attacks via AST-Guided Injectionfrom the team of Jiaheng Zhang, National University of Singapore; a new benchmark · When multiple skills are combined, malicious operations can be spread across the dependency chain, so auditing a single skill package makes it difficult to identify their combined intent.

Chosen by rule: venue tier, a ranked team, released code, a new benchmark and use by later papers; at most two per theme.

Direction radar

DirectionThis periodAverage before
↑ Agentic system architecture and runtime engineering71.7
↑ Coding agents and software engineering52.7
↑ Safety and alignment in reinforcement learning for language agents30.7
↑ Contextual integrity and privacy leakage41.7
↑ Runtime authority and injection diagnosis31
↓ Reward hacking, self-improvement, research agents05.7
↓ Embodied task-planning safety04
↓ Agent societies and cooperation03.7

Who is moving

Institutions with their first paper in the field: Kontext, Nanchang Hangkong University, Affiliation:, Auburn University at Montgomery, Belleville, CUHK (Shenzhen), Charles Sturt University, Conversational Technologies, Cookeville, Fraunhofer Institute for Molecular Biology and Applied Ecology

Open source and models

Which conclusions changed

Papers on the site5,645 → 5,781
New papers in the last 12 months3,773 → 3,868
Institutions with a paper on the site3,585 → 3,646
Open-source repositories in the library7,724 → 7,792

The other 9 tracked numbers moved by less than half a point.

Revision of the period before

2026-10-04 – 2026-10-10 was first counted at 51 papers; 26 arrived later, 77 now.

Open this issue in the Astrolabe