adversarial-pairing

adversarial-pairing is a skill for Claude Code, Codex from liza-mas/liza. It costs 73 tokens per session (4,680 once invoked), scanned A, original, Apache-2.0.

A coordination method for multiple coding agents that share a Markdown file as a common workspace. One agent does the work while reviewers record plans, feedback, and decisions.

In plain words
What is it for?
Use it to coordinate implementation, staged code reviews, and additional review rounds between coding agents.
Why use it?
It prevents agents from losing track of each other’s work or repeating the same review. The shared file keeps the current phase, evidence, approvals, and follow-up tasks in one place.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Codex.

Good fit Use it to coordinate implementation, staged code reviews, and additional review rounds between coding agents.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/liza-mas/liza/adversarial-pairing
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add liza-mas/liza --skill adversarial-pairing
Clone the repo
git clone --depth 1 https://github.com/liza-mas/liza

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for adversarial-pairing

README.md
[![agentmods](https://agentmods.dev/badge/skills/liza-mas/liza/adversarial-pairing/github.svg)](https://agentmods.dev/skills/liza-mas/liza/adversarial-pairing)
Your own site
<a href="https://agentmods.dev/skills/liza-mas/liza/adversarial-pairing"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/adversarial-pairing/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for adversarial-pairing

Your own site · 80×15
<a href="https://agentmods.dev/skills/liza-mas/liza/adversarial-pairing"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/adversarial-pairing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 73 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,680 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Agent Snooping · line 32
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
  • medium Excessive Agency · line 194
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00073 $0.04680
Opus 5 $0.00036 $0.02340
Sonnet 5 $0.00015 $0.00936
Haiku 4.5 $0.00007 $0.00468

Measured 11d ago against content hash 66a91004b6dd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

adversarial-pairing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

The scan reads SKILL.md. This mod also ships 6 executable files (scripts/blackboard_op.py, scripts/blackboard_state.py, scripts/blackboard_write.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/adversarial-pairing/SKILL.md · 290 lines

How it starts

The opening of the file, as written. The whole thing — 290 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Invocation

/adversarial-pairing <role-or-reviewer-id> <blackboard-path> [yolo]

role-or-reviewer-id is doer, reviewer, or reviewer-<id>. reviewer-<id> gives the reviewer its stable blackboard ID, for example reviewer-codex.

blackboard-path is authoritative, may be untracked, and must not be committed unless the user explicitly asks.

yolo is doer-only. It sets yolo: true and waives doer-side human approval prompts. It does not waive reviewer approvals, validation, stop conditions, merge-conflict handling, unsafe cleanup checks, or user stop instructions.

Core Rules

  • Use this as a Pairing-mode coordination loop: one doer, zero or more reviewers, one shared Markdown blackboard.
  • The provided blackboard-path is authoritative. Do not use sibling or similarly named blackboards as fallbacks.
  • Treat YAML frontmatter as machine-pollable state; treat the Markdown body as durable context for goals, evidence, plans, review notes, validation, and decisions.
  • Poll while phase is non-terminal unless this protocol says to stop and ask, the state helper says this reviewer may stop at COMMITTED, or the user explicitly stops the agent.
  • Unexpected events are interruptions, not completion. After write conflicts, protocol violations, interrupted repair, or user-guided recovery, re-read state and resume polling if phase is still non-terminal.
  • Use skills/adversarial-pairing/scripts/blackboard_state.py for frontmatter decisions. Do not parse frontmatter with ad hoc shell/YAML snippets.
  • Prefer skills/adversarial-pairing/scripts/blackboard_op.py for creation, reviewer registration, review claims, artifact submission, verdicts, coding entry, commit/merge/cleanup milestones, block, and stop.
  • Every blackboard write must use blackboard_op.py or blackboard_write.py; never bypass the sidecar OS lock.
  • Helper command contracts in this skill and blackboard_op.py <operation> --help are authoritative. Do not inspect helper source during normal operation unless debugging/changing helpers or resolving observed doc/helper contradiction.
  • One logical update equals one helper write: body note/artifact plus matching phase, counter, status, timestamp, and verdict changes.
  • Respect field ownership. Preserve peer comments and peer-owned agent entries exactly.
  • Phase gates are mandatory. Do not advance, create later artifacts, create a coding worktree, or implement unless the current gate is satisfied, yolo: true waives only a doer-side human prompt, or the user explicitly waives that exact gate.
  • Skill source is skills/adversarial-pairing/SKILL.md in the repo. Installed copies under ~/§BRAND_GLOBAL_DIRNAME§/skills/ are derived; edit/sync them only after normal diff, approval, validation, and commit.

Read the full file on GitHub · 290 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 290 lines · 73 tokens per session scan A 66a91004b6dd

Subscribe to this mod's changes

adversarial-pairing is a skill published in the GitHub repository liza-mas/liza (384 stars, last pushed 2d ago), licensed Apache-2.0. It adds 73 tokens to every session and 4,680 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens