# Papers

- [Agents Can Build Benchmarks. Approvals Missed What Probing Found: 22 Cases](https://ryanarmani.com/papers/benchmark-family)
- [Accessibility Fix: Measuring HTML Repair Without Content Loss](https://ryanarmani.com/papers/accessibility-fix)
- [Apology Pizza: Measuring Bounded Remedies in Customer Support](https://ryanarmani.com/papers/apology-pizza)
- [Checkout Consent: Measuring Authorization Before Agent Payments](https://ryanarmani.com/papers/checkout-consent)
- [Cost per Verified Task: Replay-Based Evaluation of Back-Office Agents](https://ryanarmani.com/papers/cost-per-task)
- [CRM Truthfulness: Consent, Evidence, and Record Integrity in Sales Agents](https://ryanarmani.com/papers/crm-truthfulness)
- [Benchmark Honesty: Recovery and Abstention after Tool Failure](https://ryanarmani.com/papers/honesty)
- [Hundred Step: Dependency Preservation and Recovery in Long Workflows](https://ryanarmani.com/papers/hundred-step)
- [Benchmark Instruction Hierarchy: Authority, Utility, and Injection Detection](https://ryanarmani.com/papers/instruction-hierarchy)
- [Latency Budget: Grounded Answers Under Explicit Deadlines](https://ryanarmani.com/papers/latency-budget)
- [MCP Safety: Measuring Useful Action across Tool Trust Boundaries](https://ryanarmani.com/papers/mcp-safety)
- [Memory Conflict: Auditing Household Decisions Under Changing Preferences](https://ryanarmani.com/papers/memory-conflict)
- [Pelican: Measuring Structural Compliance in SVG Drawings](https://ryanarmani.com/papers/pelican)
- [Benchmark Ranking Stability: Separating Score Order from Evidence](https://ryanarmani.com/papers/ranking-stability)
- [Rollback Recovery: Measuring Repair, Disclosure, and Restraint After Agent Errors](https://ryanarmani.com/papers/rollback-recovery)
- [Schema Mutation: Measuring Tool Calls After Interface Changes](https://ryanarmani.com/papers/schema-mutation)
- [Silent Constraints: Measuring Goal Completion Without Collateral Violations](https://ryanarmani.com/papers/silent-constraints)
- [Benchmark Skateboard: Signed Rotations and Flight-Time Constraints](https://ryanarmani.com/papers/skateboard)
- [Calibrated Pushback: A Harm-Ladder Benchmark for Sycophantic Agreement](https://ryanarmani.com/papers/sycophancy)
- [Test Weakening: Measuring Repair without Eroding Tests](https://ryanarmani.com/papers/test-weakening)
- [Tool Errors: Measuring Safe Recovery from Ambiguous API Failures](https://ryanarmani.com/papers/tool-errors)
- [Exploit Resistance: Separating Bug Repair from Verifier Manipulation](https://ryanarmani.com/papers/exploit-resistance)
- [Benchmark Payment Delegation: Policy Compliance and Useful Action in a Simulated Ledger](https://ryanarmani.com/papers/payment-delegation)