No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by zscole · Claude Skill · ★ 557
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is adversarial-spec safe to install? View the security audit →
adversarial-spec A Claude Code plugin that iteratively refines product specifications through multi-model debate until consensus is reached. Key insight: A single LLM reviewing a spec will miss things. Multiple LLMs debating a spec will catch gaps, challenge assumptions, and surface edge cases that any one model would overlook. The result is a document that has survived rigorous adversarial review. Claude is an active participant, not just an orchestrator. Claude provides independent critiques, challenges opponent models, and contributes substantive improvements alongside external models. Quick Start How It Works You describe product -- Claude drafts spec -- Multiple LLMs critique in parallel | v | Claude synthesizes + adds own critique | v | Revise and repeat until
| Stars | 557 |
| Forks | 48 |
| Language | Python |
| Category | Claude Skill |
| License | MIT |
| Quality Score | 71.9294071814604/100 |
| Open Issues | 3 |
| Last Updated | 2026-01-22 |
| Created | 2026-01-11 |
| Platforms | claude-code, python |
| Est. Tokens | ~8k |
These tools work well together with adversarial-spec for enhanced workflows:
Looking for a adversarial-spec alternative? If you're comparing adversarial-spec with other claude skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
A curated list of Plugins that let you extend Claude Code with custom commands, agents, hooks, and MCP servers
HeyClaude (formerly Claude Pro Directory) is a searchable collection of pre-built AI skills, agents, MCP serve
A practical Claude Code guide with clear mental models and copy-paste examples — setup, prompt engineering, sl
Battle-tested Claude Code, OpenAI Codex, Cursor configs, plugins, hooks and agents with Kimi, MiniMax and GLM
A local sandbox for your AI agents
The Complete AI Development Toolkit for Claude Code. 106 skills, 36 agents, 171 hooks. Install `ork` for stabl
Explore other popular claude skill tools:
adversarial-spec is A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until all models reach consensus.. It is categorized as a Claude Skill with 557 GitHub stars.
adversarial-spec is primarily written in Python. It covers topics such as anthropic, claude-ai, claude-code.
You can find installation instructions and usage details in the adversarial-spec GitHub repository at github.com/zscole/adversarial-spec. The project has 557 stars and 48 forks, indicating an active community.
adversarial-spec is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to adversarial-spec on Agent Skills Hub include awesome-claude-plugins, claudepro-directory, Claude-Code-Everything-You-Need-to-Know. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: