Adversarial Spec Agent Skill

by zscole• 557•• Scanned•Claim this listing•Request removal
Developer Toolsanthropicclaude aiclaude codeclaude code pluginclaude skills

A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until all models reach consensus.

Adversarial Spec is an open-source agent skill by zscole, written mainly in Python, with 557 GitHub stars and the MIT license. Its code was last updated Jan 22, 2026.

Scan summary: scanned Nov 5, 2026 at commit f90cf0c. Seen able to connect to the internet. Scan policy

What Adversarial Spec can touch

  • Run shell commandsNot seen in this scan
  • Run dynamic codeNot seen in this scan
  • Read or write filesNot seen in this scan
  • Read environment and secretsNot seen in this scan
  • Control a web browserNot seen in this scan
  • Connect to databasesNot seen in this scan
  • Read or send messagesNot seen in this scan
  • Touch money or tradingNot seen in this scan
  • Connect to the internetFound in the code by our scan

"Not seen" means our checks did not find it, and the tool may still do it. Covers commit f90cf0c, Nov 5, 2026. See the scan policy.

Before you install Adversarial Spec

AgenticMart scanned Adversarial Spec on Nov 5, 2026 with automated checks for hidden code, known-malicious packages, leaked secrets and vulnerable dependencies. The result covers one commit under our scan policy, so review what access you give it and check any advisories shown here.

What it can access

  • A skill is instructions plus optional scripts. The instructions steer your agent; any scripts run on your machine when the agent uses the skill.

Before you install

  1. Pin the version you reviewed instead of always installing the latest release.
  2. Read the tool list or scripts before the first run, and keep approval prompts on for actions that write, send or delete.
  • Maintained by zscole on GitHub, 557 stars and 51 forks. Popularity is not proof of safety.
  • Last code change: Jan 22, 2026.
  • Open-source license: MIT.

Adversarial Spec is an open-source agent skill by zscole, written mainly in Python. A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until all models reach consensus.

At a glance

Stars
557
Forks
51
Language
Python
License
MIT
Created
Jan 2026
Last update
Jan 22, 2026
Type
Agent Skill
Price
Free

Topics

anthropicclaude-aiclaude-codeclaude-code-pluginclaude-skillsllmorchestration

Questions about Adversarial Spec

What is Adversarial Spec?

A Claude Code plugin that iteratively refines product specifications by debating between multiple LLMs until all models reach consensus.

Who maintains Adversarial Spec?

Adversarial Spec is maintained by zscole on GitHub (zscole/adversarial-spec). AgenticMart lists it from public repository data.

Is Adversarial Spec safe to use?

AgenticMart scanned Adversarial Spec on Nov 5, 2026 with automated checks for hidden code, known-malicious packages, leaked secrets and vulnerable dependencies. The result covers one commit under our scan policy, so review what access you give it and check any advisories shown here.

What license does Adversarial Spec use?

It is published under the MIT license.

Adversarial Spec alternatives

View more →

Similar agent skills in Developer Tools, ranked by GitHub stars. Compare licence, recent activity and scan status before you choose.

NameStarsLicenceLast updatedScan
Adversarial Spec (this page)
zscole
557MITJan 22, 2026Scanned, no flagged issues
Claude Skills
alirezarezvani
29kMITOct 31, 2026Scanned, no flagged issues
Pm Skills
phuryn
28kMITOct 29, 2026Scanned, no flagged issues
Notebooklm Py
teng-lin
20kMITOct 31, 2026Scanned, no flagged issues
Openrig
mvschwarz
8.5kApache-2.0Nov 1, 2026Scanned, no flagged issues
Yao
YaoApp
8.2kNone foundOct 19, 2026Not scanned yet