Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks

• URL: http://arxiv.org/abs/2605.18583v1
• Date: Mon, 18 May 2026 16:00:41 GMT
• Title: Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks
• Abstract summary: OverEager-Gen is a benchmark dedicated to overeager behavior on benign tasks.<n>On Claude Code, stripping the consent declaration alone raises the overeager rate from 0.0% to 17.1%.<n>OverEager-Gen therefore certifies each scenario's discriminative power before admission.
• Score: 40.270213696031625
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Coding agents now run autonomously with shell, file, and network privileges. When a user issues a benign request, the agent sometimes does more than asked: it deletes unrelated files, wipes a stale credentials backup, or rewrites configuration the user never mentioned. We call these scope expansions overeager actions, an authorization problem distinct from capability failures, prompt injection, or sandbox escapes. We present OverEager-Gen, a benchmark dedicated to overeager behavior on benign tasks. Building it surfaces a measurement-validity issue: if a benchmark spells out the authorized scope inside the prompt, the agent stops inferring boundaries and starts pattern-matching declaration text. On Claude Code, stripping the consent declaration alone raises the overeager rate from 0.0% to 17.1% on paired scenarios (McNemar exact p = 2.4 x 10^-4). OverEager-Gen therefore certifies each scenario's discriminative power before admission via a behavioral-gradient validator, audits internal tool calls through a dual-channel stack (PATH-injected shim plus per-agent event streams), and ships byte-identical consent_kept and consent_stripped variants. OverEager-Bench contains 500 validated scenarios and ~7,500 runs across four agent products (Claude Code, OpenHands, Codex CLI, Gemini CLI) and six base models; a 50-sample re-annotation gives Cohen's kappa = 0.73 and rule-judge recall = 1.00. Stripping consent multiplies the overeager rate on every shared base model (Delta in [11.9, 17.2] pp). The framework axis dominates effect size: a permissive cluster (Claude Code, Codex CLI, Gemini CLI) runs at 5.4-27.7% while the ask-to-continue framework (OpenHands) sits at 0.2-4.5% (Fisher p <= 10^-5). Within-framework base-model variance reaches 15.9 pp, indicating that model-layer alignment does not fully propagate through permissive permission gating.

Related papers

• Calibration is the Bottleneck: An Action-Class Diagnostic of Multi-Turn Tool-Calling [65.03665015319457]
  We propose an action-class-oriented diagnostic framework that decomposes multi-turn failures into two modes: action-class miscalibration and action-execution failure.<n>We validate it on a panel of tool-calling models across multiple multi-turn benchmarks.
  arXiv  Detail & Related papers  (2026-09-01T09:08:15Z)
• ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents [61.184954205350984]
  ClawArena-Team is a benchmark of 41 multi-turn, multimodal, multi-agent scenarios spanning 258 evaluation rounds and 72 staged updates.<n>All scoring is execution-based with no LLM judge: an overall score -- the Subagent-Management Score (SMS) -- multiplies task correctness by a least-privilege and modality-routing factor.
  arXiv  Detail & Related papers  (2026-06-30T06:06:08Z)
• Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens [65.53494487819053]
  Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services.<n>We adopt a computer-system analogy: treating a Claw-like agent as an agentic computer system whose gateway runtime plays an OS-like mediation role.<n>We develop SafeClawArena, a benchmark of 406 adversarial tasks across four attack surfaces.
  arXiv  Detail & Related papers  (2026-06-29T18:00:45Z)
• How Reliable Are AI Attackers Against a Fixed Vulnerable Target? A 400-Run Empirical Study of LLM Penetration Testing Consistency [0.0]
  Large language models (LLMs) can autonomously conduct multi-stage cyber attacks, but the consistency of their offensive behavior under repeated penetration unstudied.<n>This work presents the first large-scale empirical measurement of attack consistency: 400 runs (4 models, 100 each) against an identical honeypot hosting Juice Shop.<n>No model emitted a content refusal that survived the orchestrator's one-shot authorization re-prompt at 0-1.
  arXiv  Detail & Related papers  (2026-05-28T15:39:43Z)
• SNARE: Adaptive Scenario Synthesis for Eliciting Overeager Behavior in Coding Agents [40.270213696031625]
  A coding agent executes a benign task as a sequence of shell, file, and network actions, any of which can quietly exceed the authorized scope while the task still completes.<n>We call this overeager behavior: the prompt is not adversarial and the run succeeds, yet an out-of-scope step can leak credentials or delete files.<n>Existing benchmarks miss it: task-completion suites credit any finished run, jailbreak suites probe adversarial prompts, and the one prior overeager benchmark applies a single fixed prompt to every agent-model pair.<n>We present SNARE, a pipeline that composes benign scenarios from reusable scope
  arXiv  Detail & Related papers  (2026-05-27T08:14:07Z)
• WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation [88.10947115397971]
  This work presents WildClawBench, a native-runtime benchmark of 60 human-authored, bilingual, multimodal tasks spanning six thematic categories.<n>Each task averages roughly 8 minutes of wall-clock time and over 20 tool calls.<n> Grading is hybrid, combining deterministic rule-based checks, environment-state auditing of side effects, and an LLM/VLM judge for semantic verification.
  arXiv  Detail & Related papers  (2026-05-11T17:49:43Z)
• Instruction Adherence in Coding Agent Configuration Files: A Factorial Study of Four File-Structure Variables [0.0]
  coding agents read configuration files at session start and are expected to follow the conventions inside them.<n>We report a systematic factorial study of these choices using four manipulated variables.<n>None of the four structural variables or three two-way interactions produces a detectable contrast after multiple-testing correction.
  arXiv  Detail & Related papers  (2026-05-11T06:09:47Z)
• MOSAIC-Bench: Measuring Compositional Vulnerability Induction in Coding Agents [2.1942030377331245]
  Coding agents often pass per-prompt safety review yet ship exploitable code when their tasks are decomposed into routine engineering tickets.<n>We introduce MOSAIC-Bench, a benchmark of 199 three-stage attack chains paired with deterministic exploit oracles on deployed software substrates.<n>We show that nine production coding agents compose innocuous tickets at 53-86% end-to-end ASR with only two refusals across all staged runs.
  arXiv  Detail & Related papers  (2026-05-05T16:38:23Z)
• ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents [77.22389710754452]
  We introduce bench, a benchmark for coworker agents built around multi-turn multi-day tasks.<n>The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services.<n>The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0%, indicating that partial progress is common.
  arXiv  Detail & Related papers  (2026-04-26T16:05:02Z)
• ClawArena: Benchmarking AI Agents in Evolving Information Environments [61.664633997138004]
  ClawArena is a benchmark for evaluating AI agents in evolving information environments.<n>Each scenario maintains a complete hidden ground truth while exposing the agent only to noisy, partial, and sometimes contradictory traces.<n> evaluation is organized around three coupled challenges: multi-source conflict reasoning, dynamic belief revision, and implicit personalization.
  arXiv  Detail & Related papers  (2026-04-05T17:55:23Z)
• Governing Dynamic Capabilities: Cryptographic Binding and Reproducibility Verification for AI Agent Tool Use [0.0]
  No existing security layer verifies what an AI agent can do, whether it executed what it claims, or what happened in a multi-agent interaction.<n>Existing frameworks conflate the two, enabling silent capability escalation and leaving interactions without verifiable provenance.<n>We derive three Agent Governance Requirements: capability integrity (G1), behavioral verifiability (G2), and interaction auditability (G3)<n>We validate with two crypto-agnostic instantiations -- basic (Ed25519, SHA-256; 97 us verify) and enhanced (BBS+ selective disclosure, Groth16 DV-SNARK; 13.8 ms)
  arXiv  Detail & Related papers  (2026-03-15T11:46:57Z)
• OpenSec: Measuring Incident Response Agent Calibration Under Adversarial Evidence [0.0]
  We introduce OpenSec, a dual-control reinforcement learning environment that evaluates defensive incident response agents.<n>Unlike static capability benchmarks, OpenSec scores world-state-changing containment actions under adversarial evidence.<n>We find consistent over-triggering in this setting: GPT-5.2, Gemini 3, and DeepSeek execute containment in 100% of episodes with 90-97% false positive rates.
  arXiv  Detail & Related papers  (2026-01-28T22:12:54Z)
• SOPBench: Evaluating Language Agents at Following Standard Operating Procedures and Constraints [59.645885492637845]
  SOPBench is an evaluation pipeline that transforms each service-specific SOP code program into a directed graph of executable functions.<n>Our approach transforms each service-specific SOP code program into a directed graph of executable functions and requires agents to call these functions based on natural language SOP descriptions.<n>We evaluate 18 leading models, and results show the task is challenging even for top-tier models.
  arXiv  Detail & Related papers  (2025-03-11T17:53:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.