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.
npx skills add karsten-s-nielsen/mad-scientist-skills --skill cognitive-interface-auditgit clone --depth 1 https://github.com/karsten-s-nielsen/mad-scientist-skillsWrote 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.
[](https://agentmods.dev/skills/karsten-s-nielsen/mad-scientist-skills/cognitive-interface-audit)<a href="https://agentmods.dev/skills/karsten-s-nielsen/mad-scientist-skills/cognitive-interface-audit"><img src="https://agentmods.dev/badge/skills/karsten-s-nielsen/mad-scientist-skills/cognitive-interface-audit/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.
<a href="https://agentmods.dev/skills/karsten-s-nielsen/mad-scientist-skills/cognitive-interface-audit"><img src="https://agentmods.dev/badge/skills/karsten-s-nielsen/mad-scientist-skills/cognitive-interface-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00230 | $0.17774 |
| Opus 5 | $0.00115 | $0.08887 |
| Sonnet 5 | $0.00046 | $0.03555 |
| Haiku 4.5 | $0.00023 | $0.01777 |
Grade A, and why
cognitive-interface-audit 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.
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.
How it starts
The opening of the file, as written. The whole thing — 866 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cognitive Interface Audit
A comprehensive cognitive interface audit with two modes and a single tier:
Modes:
- Planning (before UI exists) — task model design, user expertise mapping, error tolerance strategy, information architecture
- Audit (on existing UI) — mental model alignment, consistency, error tolerance, cognitive load, visual grounding, accessibility
Single tier: Unlike security auditing, cognitive interface evaluation is methodology-based (GOMS analysis, heuristic evaluation, NASA-TLX scoring), not tool-licensed. No paid software is required. The value is in the analytical framework, not the scanner.
Core question: "Does the interface think the way the user thinks?"
Academic foundations
This skill synthesizes seven research threads into a single audit methodology:
-
Task Model & Error Tolerance (Card, Moran & Newell 1983; Wood & Byrne 2002; Rasmussen 1983) — GOMS models predict how users decompose tasks. Wood's 7-layer defense framework predicts where errors occur and what defenses are needed at each stage. Rasmussen's SRK framework classifies errors by cognitive level (skill-based slips, rule-based misapplication, knowledge-based wrong mental model), each requiring different design countermeasures.
-
Visual Grounding & Common Ground (Gergle, Kraut & Fussell 2004/2013; Gergle et al. 2021) — Shared visual information affects task performance through two distinct mechanisms: situation awareness (does the user understand system state?) and conversational grounding (does the interface provide enough shared context?). Not just availability but the form of visual information differentially affects performance. The Joint Action Storyboard framework maps each interaction to its grounding cost.
-
Cognitive Load (Sweller 1988; Hart & Staveland 1988; Kahneman 2011) — Every interface decision either consumes or conserves working memory. NASA-TLX provides structured evaluation across 6 dimensions. Sweller's distinction between intrinsic load (inherent task complexity), extraneous load (poor design), and germane load (productive learning) guides where to invest and where to cut. Kahneman's Dual-Process Theory (System 1/System 2) identifies where interfaces force slow, deliberate processing when fast, intuitive processing could suffice.
What ships with it
5 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.
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.
- 11d ago First seen · 866 lines · 230 tokens per session scan A a5400c18ca44
cognitive-interface-audit is a skill published in the GitHub repository karsten-s-nielsen/mad-scientist-skills (3 stars, last pushed 13d ago), licensed MIT. It adds 230 tokens to every session and 17,774 once invoked, about $0.0011 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 2026-08-31.
Other skills, from other repositories
recipe-front-design
Execute from repository evidence through applicable UI Spec and optional ADR decisions to complete frontend Design Doc approval.
recipe-front-adjust
Adjust an already-implemented UI in-session with verification against the design source.
fidelity-gate
Build a UI against a frozen visual reference without drift - an inventory extracted before any code, a relics list, and a gate that MEASURES computed styles on a fixture carrying the reference's own data. Use when a mockup, design spec or screenshot is the contract.
a11y-audit
Dedicated WCAG 2.2 AA/AAA accessibility audit across 10 dimensions (A1-A10) covering semantic HTML, keyboard navigation, ARIA patterns, color contrast, forms, images/media, responsive/zoom, motion/animation, reading/content, and legal compliance. Goes far beyond surface-level design-review checks with deep…
design-review
UI/UX design consistency audit. Code-based DX1-DX20 checklist covering states, consistency, accessibility, responsive behavior, and interaction patterns. Optional visual audit via chrome-devtools screenshots and automated WCAG accessibility via axe-core. DAP1-DAP12 anti-pattern detection. Modes: [path], visual…
design
Intent-first UI design with conscious, traceable design decisions. Persists decisions in .interface-design/ for cross-session consistency. Includes domain exploration, design system generation (system.md + system.json), per-component construction with mandatory checkpoints, and craft validation tests. Modes: init…