Dynamic tool generation empowers Large Language Model (LLM) agents to synthesize tools on demand, yet a critical challenge remains: 32.4% of generated tools fail on first invocation.We present Causal Tool Diagnosis (CTD), a principled framework that moves beyond blackbox reliability prediction to interpretable failure attribution.CTD constructs a Structural Causal Model (SCM) capturing how specification quality, code characteristics, and execution environment jointly determine tool outcomes.Uniquely leveraging code's intervenability, we conduct controlled sandbox experiments to estimate causal effects-an advantage unavailable in pure text generation.CTD jointly predicts confidence (Spearman rank correlation coefficient ρ=0.90) and root cause attribution (78% accuracy), with attributions directly guiding targeted repairs (+9.6% success rate over error-type classification).Our ARCHI-TECT framework, integrating CTD throughout the tool lifecycle, achieves state-of-the-art on four benchmarks including StableToolBench (+3.8%),MINT (+4.6%),T-Eval (+3.7%), and SWE-bench Lite (+2.4%), with consistent improvements across all settings.

Large language model (LLM) agents excel at multi-step tasks yet frequently exhibit Action Boundary Blindness-the inability to correctly determine action granularity, scope, and completeness.Grounded in Event Segmentation Theory from cognitive science, we formalize three violation types: granularity confusion, scope creep, and boundary ambiguity.We propose four automatic metrics-Action Boundary Score (ABS), Granularity Alignment Rate (GAR), Scope Violation Rate (SVR), and Boundary-Aware Success Rate (BASR)requiring no human annotation.Experiments on 1,655 tasks across six benchmarks (τ -bench, WebArena, ALFWorld, TheAgentCompany, OSWorld) with seven LLMs reveal that: (1) the best model achieves only 0.424 ABS; (2) using a multi-label attribution framework validated by inter-annotator agreement (κ = 0.78), boundary blindness is the primary failure mode in 37.2% of failures (25.8% as sole cause; 55.9% total involvement including contributing factors); (3) under-action dominates at 48.4%; (4) BASR is consistently ∼4 points lower than traditional success rate, exposing "lucky successes."Critically, Explicit Boundary Prompting (EBP) improves ABS by 0.08-0.13across all models, demonstrating that boundary blindness is better characterized as an elicitation gap rather than a fundamental capability limitation-LLMs possess latent boundary perception not activated by default.This finding has implications for alignment and instruction tuning.We validate metrics through state-based cross-validation and human audit, estimating ∼22% false positive rate from valid alternative paths, with model rankings remaining stable (Spearman ρ = 1.0).