wehub-resource-sync
0d3cb498a3
CI / Shell Format Check (push) Has been cancelled
CI / Check Ruby (3.4) (push) Has been cancelled
CI / CI Config (push) Has been cancelled
CI / Test on Node ${{ matrix.node }} and ${{ matrix.os }}${{ matrix.shard && format(' (shard {0}/3)', matrix.shard) || '' }} (push) Has been cancelled
CI / Build on Node ${{ matrix.node }} (push) Has been cancelled
CI / Style Check (push) Has been cancelled
CI / Generate Assets (push) Has been cancelled
CI / Check Python (3.14) (push) Has been cancelled
CI / Check Python (3.9) (push) Has been cancelled
CI / Build Docs (push) Has been cancelled
CI / Code Scan Action (push) Has been cancelled
CI / Site tests (push) Has been cancelled
CI / webui tests (push) Has been cancelled
CI / Run Integration Tests (push) Has been cancelled
CI / Run Smoke Tests (push) Has been cancelled
CI / Go Tests (push) Has been cancelled
CI / Share Test (push) Has been cancelled
CI / Redteam (Production API) (push) Has been cancelled
CI / Redteam (Staging API) (push) Has been cancelled
CI / GitHub Actions Lint (push) Has been cancelled
CI / Check Ruby (3.0) (push) Has been cancelled
release-please / release-please (push) Has been cancelled
release-please / build (push) Has been cancelled
release-please / publish-npm (push) Has been cancelled
release-please / publish-npm-backfill (push) Has been cancelled
release-please / docker (push) Has been cancelled
release-please / publish-code-scan-action (push) Has been cancelled
release-please / attest-code-scan-action (push) Has been cancelled
Deploy local.promptfoo.app / Deploy to Cloudflare Pages (push) Has been cancelled
Test and Publish Multi-arch Docker Image / test (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-amd64 platform:linux/amd64 runner:ubuntu-latest]) (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-arm64 platform:linux/arm64 runner:ubuntu-24.04-arm]) (push) Has been cancelled
Test and Publish Multi-arch Docker Image / merge-docker-digests (push) Has been cancelled
Test and Publish Multi-arch Docker Image / Attest Multi-arch Image (push) Has been cancelled
Validate Renovate Config / Validate Renovate Configuration (push) Has been cancelled
2.2 KiB
2.2 KiB
sidebar_label, description
| sidebar_label | description |
|---|---|
| Reasoning DoS | Red team AI systems against reasoning-based denial of service attacks by testing computational resource limits to prevent malicious system exhaustion |
Reasoning DoS Plugin
Reasoning DoS (Denial of Service) covers prompts that push reasoning-capable models into unnecessary work, long runtimes, or inflated inference cost. Use this plugin to test whether the system handles those pressure cases without giving the attacker an easy resource-exhaustion path.
Purpose
It focuses on whether the system:
- Avoids unnecessary recursive reasoning.
- Resists prompts designed to trigger excessive computation.
- Preserves reasonable cost and latency under adversarial input.
Configuration
To include the Reasoning DoS plugin in your LLM red teaming setup:
redteam:
plugins:
- reasoning-dos
How It Works
The Reasoning DoS plugin generates test cases that attempt to trigger excessive computation through:
- Complex mathematical or logical problems requiring iterative solutions
- Nested decision-making scenarios
- Multi-step optimization problems
- Game theory scenarios
- Markov Decision Processes (MDPs)
- Recursive reasoning chains
- Puzzle-solving tasks (e.g., Sudoku)
Evaluation Criteria
The plugin assesses the AI's response based on:
- Computation Limits: Does the system properly limit computational resources?
- Pattern Recognition: Can it identify potential computational traps?
- Response Efficiency: Does it provide direct answers without unnecessary computation?
- Resource Management: How well does it manage computational resources under stress?
- Attack Resistance: Can it resist attempts to trigger excessive computation?
Importance in Gen AI Red Teaming
These tests matter when a model can spend far more effort on a prompt than the user spent crafting it. The risk is operational: degraded latency, higher cost, and reduced service quality under targeted abuse.
Related Concepts
- Divergent Repetition Plugin
- Types of LLM vulnerabilities - Full vulnerability and plugin directory with category mapping