yzj623/pplab/adaptive-apgd-loop
Claude Code and agent skill by yzj623/pplab
- community
Checking that the SKILL.md file is still in the repository.
Run a deterministic APGD-style projected attack with loss-history step adaptation and auditable traces.
Install
npx vskill@latest install yzj623/pplab/adaptive-apgd-loopSKILL.md
The same SKILL.md was published earlier as vectorspacelab/arex-skill/adaptive-apgd-loop.
Adaptive APGD Loop
Use this skill to recover or test the APGD mechanism from AutoAttack when a bounded, deterministic implementation is needed. Do not treat this pure-Python finite-difference implementation as a high-throughput replacement for framework autograd attacks.
Inputs
- Logit function accepting one vector.
- Examples and labels.
- Projection and loss helpers.
- Norm, epsilon, iterations, initial step size, and adaptation window.
Outputs
- Adversarial examples and predictions.
- Per-example success flags.
- Loss trace and step-size adaptation events.
Workflow
- Start from the clean example or a deterministic offset inside the threat ball.
- Estimate the loss gradient with finite differences.
- Take an ascent step and project back to the threat model.
- Track the best loss and adversarial point.
- On non-improving windows, halve step size and continue from the best point.
Validation
Run python tests/test_apgd.py or the skill-tree validator.
Limitations
Finite differences are only for compact recovery evidence. Production use should substitute exact gradients while preserving the same adaptation and projection contracts.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingbasicagent-skill/skill_library/paperbench/robust-clip/skill/autoattack_apgd/adaptive_apgd_loop/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/yzj623/pplab/adaptive-apgd-loop)