Paperthin is a collection of low-level design patterns that coding agents can apply as repeatable checks and actions during software work. It is for people using agents such as Claude Code, Codex, Cursor, or other supported tools, and its catalogue entries package the patterns as skills, an instruction, and a plugin.
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 LilMGenius/paperthin --skill mandelagit clone --depth 1 https://github.com/LilMGenius/paperthinWrote 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/lilmgenius/paperthin/mandela)<a href="https://agentmods.dev/skills/lilmgenius/paperthin/mandela"><img src="https://agentmods.dev/badge/skills/lilmgenius/paperthin/mandela/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/lilmgenius/paperthin/mandela"><img src="https://agentmods.dev/badge/skills/lilmgenius/paperthin/mandela.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Socket pass
- Snyk pass
- NVIDIA SkillSpector pass
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 | $0.00106 | $0.00711 |
| Opus 5.5 | $0.00042 | $0.00284 |
| Sonnet 5.5 | $0.00021 | $0.00142 |
| Haiku 4.5 | $0.00011 | $0.00071 |
Grade A, and why
mandela 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 1mo 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- mandela — 100% identical
How it starts
The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit a validation for leakage: does outside ground-truth actually enter, or is everyone confirming a result no one independently produced?
Goal
The name is the Mandela Effect — a whole population confidently remembers something that never independently happened; a leaky validation is the same shape. Walk the 8 patterns below. mandela checks one thing: whether a validation is independent, or whether the designer, model, and scorer are only confirming each other.
Workflow
- Identify the validation (eval / metric / experiment / holdout / "how we'll know"). Name its components — what plays model, scorer, designer, dataset.
- Ask the core question: does external ground-truth enter independently?
- Test the validation against all 8 patterns below (some apply only to certain components — a human subject, a scorer); report only the ones that fire, each by name.
- Give the independent-ground-truth fix for each hit.
The 8 leakage patterns
- Recall, not reason — a memorized answer recited instead of one actually derived; the system already knows the result it is supposedly computing.
- Wrong null hypothesis — an ablation that removes a surface label but not the underlying signal the system actually exploits, so the "control" still leaks.
- Shared hallucination — two components verifying each other; circularity reported as a number.
- Tautology — a scorer grading buckets it drew itself.
- Verifier = designer — a private, unreproducible recipe in a holdout's clothes.
- Shared-pool bias — train and holdout drawn from one labeler pool, so one bias enters both sides.
- Frame injection — a question that hands the subject the hypothesis.
- Demand characteristics — measured subjects who know they're being measured.
Rules
- Subtlety that bites twice: you can blind the output value and still leak the collection recipe.
- Read-only — name the leak and the independence fix; don't rewrite the experiment.
- For a high-stakes validation, you may add one independent fresh-context auditor (
N=1) handed only the validation design, not this session's reasoning, to re-run the 8-pattern taxonomy blind; the default remains same-session and read-only.
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.
- 1mo ago First seen · 43 lines · 106 tokens per session scan A 5efc42ffa98a
mandela is a skill published in the GitHub repository LilMGenius/paperthin (1,351 stars, last pushed today), licensed MIT. It adds 106 tokens to every session and 711 once invoked, about $0.0004 per session on Opus 5.5. 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.
Other skills, from other repositories
research-extraction
Deep extraction of research corpora into structured knowledge documents using a Workflow-driven subagent/reviewer pattern. Use when you have raw source documents (theses, papers, reports) and need structured extraction across multiple axes (architecture, messages, algorithms, forms, etc.) with confidence-annotated…
LQF_Machine_Learning_Expert_Guide
LQF Machine Learning Expert Guide - Routed skill for ML/Statistical Modeling with Critical Discussion Mode. Triggers on: machine learning, modeling, prediction, training, classification, regression, clustering, deep learning, neural network, model evaluation, feature engineering, hyperparameter tuning, overfitting…
paper-2-web
Use when converting academic papers into dissemination assets such as Paper2Web websites, Paper2Video video abstracts, or Paper2Poster conference posters from LaTeX or PDF sources.
plotly
Interactive scientific and statistical data visualization library for Python. Use when creating charts, plots, or visualizations including scatter plots, line charts, bar charts, heatmaps, 3D plots, geographic maps, statistical distributions, financial charts, and dashboards. Supports both quick visualizations (Plotly…
performing-causal-analysis
Estimate causal effects from existing data. Use when fitting or interpreting DiD, ITS, synthetic control, regression discontinuity, or other treatment-effect analyses, including robustness checks and counterfactual plots. For choosing a study design before analysis, use designing-experiments instead.
evo-memory
Manages persistent research memory across ideation and experimentation cycles. Maintains two stores: Ideation Memory MI (feasible/unsuccessful directions) and Experimentation Memory ME (reusable strategies for data processing, model training, architecture, debugging). Three evolution mechanisms: IDE (after…