audit-interaction

audit-interaction is a command for Claude Code from Owl-Listener/ai-design-skills. It costs 11 tokens per session (454 once invoked), scanned A, original, MIT.

A review method for an existing human-AI interaction, meaning the way a person and an AI take turns and work together. It checks the interaction against established patterns for conversation and cooperation.

In plain words
What is it for?
Reviewing chat flows, classifying how an interaction works, rating turn-taking, checking initiative balance, and finding repair mechanisms.
Why use it?
It helps reveal confusing turn-taking, poor control over who leads, and weak ways to correct misunderstandings.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the model-interaction-design plugin — 8 skills, 3 commands shipped together

Good fit Reviewing chat flows, classifying how an interaction works, rating turn-taking, checking initiative balance, and finding repair mechanisms.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/owl-listener/ai-design-skills/audit-interaction
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.

Clone the repo
git clone --depth 1 https://github.com/Owl-Listener/ai-design-skills

Made for: Claude Code.

Or install model-interaction-design, the plugin that ships this one along with the rest of its 8 skills, 3 commands.

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 audit-interaction

README.md
[![agentmods](https://agentmods.dev/badge/commands/owl-listener/ai-design-skills/audit-interaction/github.svg)](https://agentmods.dev/commands/owl-listener/ai-design-skills/audit-interaction)
Your own site
<a href="https://agentmods.dev/commands/owl-listener/ai-design-skills/audit-interaction"><img src="https://agentmods.dev/badge/commands/owl-listener/ai-design-skills/audit-interaction/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 audit-interaction

Your own site · 80×15
<a href="https://agentmods.dev/commands/owl-listener/ai-design-skills/audit-interaction"><img src="https://agentmods.dev/badge/commands/owl-listener/ai-design-skills/audit-interaction.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 454 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.
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 $0.00011 $0.00454
Opus 5 $0.00005 $0.00227
Sonnet 5 $0.00002 $0.00091
Haiku 4.5 $0.00001 $0.00045

Measured 10d ago against content hash 1e452ab846ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

audit-interaction 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 10d ago.

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.

claude-plugin/model-interaction-design/commands/audit-interaction.md · 46 lines

What it actually says

You are auditing an existing human-AI interaction. Use only skills from the model-interaction-design plugin. Follow this process:

Step 1: Classify the Interaction Mode

Using conversation-patterns, identify:

  • What dialogue structure is being used? (interview, co-creation, instruction-execution, exploration, guided workflow)
  • Is this the right structure for the task?
  • Are there moments where the structure breaks down?

Step 2: Evaluate Turn-Taking

Using conversation-patterns:

  • Are turns appropriately sized for the task?
  • Does the AI know when to stop talking?
  • Are there awkward turn boundaries?
  • Rate turn-taking quality (1-5) with justification

Step 3: Assess Initiative Balance

Using mixed-initiative-flow:

  • Who leads at each stage? Is this appropriate?
  • Are handoffs clean or confusing?
  • Does the AI take initiative when it should? Hold back when it should?
  • Are there initiative anti-patterns (whiplash, passive AI, overbearing AI)?

Step 4: Check Repair Mechanisms

Using conversation-patterns and feedback-loops:

  • When misunderstandings happen, how are they repaired?
  • Can the user correct the AI easily?
  • Does the AI acknowledge and recover from errors?
  • Are there grounding checkpoints?

Step 5: Evaluate Disclosure

Using progressive-disclosure:

  • Does the user know what the AI can do?
  • Are capabilities revealed at the right pace?
  • Is there evidence of underuse (user doesn't know about features)?

Step 6: Review Context Design

Using context-window-design:

  • Does the AI maintain context across the conversation?
  • Are there moments where context is lost?
  • Is memory handled gracefully?

Output

Deliver an interaction audit report:

  1. Overall quality score (1-5) with justification
  2. Findings table: Issue | Severity | Skill Area | Recommendation
  3. Top 3 improvements ranked by impact
  4. Revised conversation flow showing recommended changes
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. 10d ago First seen · 46 lines · 11 tokens per session scan A 1e452ab846ba

Subscribe to this mod's changes

audit-interaction is a command published in the GitHub repository Owl-Listener/ai-design-skills (172 stars, last pushed 3mo ago), licensed MIT. It adds 11 tokens to every session and 454 once invoked, about $0.0001 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.