New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Topic · AI & LLM Engineering

Best AI interpretability skills for Claude Code, Codex and other agents.

Skills that analyse what models represent internally and how to steer them.
skills
23
official
1

AI interpretability skills, ranked

Ranked by score. Sort bymost stars,trending,newest,recently updated

AI interpretability skills, ranked
#SkillRepositoryStarsUsed inTokensAuto-checkLicenceUpdated
1
1.Esmfold2

Biohub ESMFold2 / ESMFold2-Fast all-atom co-folding (Candido et al.

JimLiu/science-skills2274 repos~2.5kAutomated safety check: PassApache-2.03 mo ago
2
2.Obliteratus

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails…

RedWoodOG/Hermes-Desktop1776 repos~3.8kAutomated safety check: PassMIT4 mo ago
3
3.Anomalib Adding A Model

Adds a new anomaly-detection model to anomalib under src/anomalib/models/.

open-edge-platform/anomalib6.2k—~1.9kAutomated safety check: PassApache-2.0today
4
4.Sae Feature Annotations

Look up what a Biohub ESM-C sparse-autoencoder (SAE) feature means — its label, description, top-activating proteins, decoder neighbours, and activation statistics — by querying the Biohub…

softnanolab/bagel148—~1.5kAutomated safety check: PassMITtoday
5
5.Sparse Autoencoder Training with SAELens

Guides training and analyzing sparse autoencoders with SAELens to break neural network activations into interpretable features, including superposition and monosemanticity studies.

Orchestra-Research/AI-Research-SKILLs13k6 repos~3.2kAutomated safety check: PassMIT3 mo ago
6
6.TransformerLens Interpretability

Guides mechanistic interpretability work with TransformerLens: loading models, caching activations, using HookPoints, activation patching and attention-pattern analysis.

Orchestra-Research/AI-Research-SKILLs13k4 repos~3kAutomated safety check: PassMIT3 mo ago
7
7.Nnsight Remote Interpretability

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution.

Orchestra-Research/AI-Research-SKILLs13k3 repos~3.3kAutomated safety check: PassMIT3 mo ago
8
8.pyvene Causal Interventions

Guides causal experiments on PyTorch models with pyvene, such as causal tracing, activation patching and interchange intervention training, to test how a model works.

Orchestra-Research/AI-Research-SKILLs13k3 repos~3.5kAutomated safety check: PassMIT3 mo ago
9
9.SHAP Model Explainability

Explains machine learning predictions with SHAP: picking the right explainer, computing Shapley values and drawing waterfall, beeswarm, bar and force plots.

davila7/claude-code-templates32k12 repos~4.6kAutomated safety check: PassMITtoday
10
10.Corvus Buffer And Pooling

Write allocation-efficient buffer code in Corvus.JsonSchema using the codebase's established three-tier pooling pattern: stackalloc → ArrayPool → ThreadStatic caches.

corvus-dotnet/Corvus.JsonSchema199—~2.9kAutomated safety check: PassApache-2.0today
11
11.Geoffrey Hinton

Applies the reasoning style of Geoffrey Hinton, deep learning pioneer and 2018 Turing Award winner.

K-Dense-AI/mimeo282—~1.8kAutomated safety check: PassMIT1 mo ago
12
12.Bias Detection Design

Designing review workflows to surface and mitigate bias in AI outputs.

Owl-Listener/ai-design-skills180—~650Automated safety check: PassMIT3 mo ago
13
13.Richard S Sutton

Reach for this skill whenever you are discussing reinforcement learning, agentic AI systems, AI alignment, continual learning, or the philosophical limits of large language models.

K-Dense-AI/mimeo282—~1.8kAutomated safety check: PassMIT1 mo ago
14
14.Saelens

Train sparse autoencoders to interpret model features. An agent skill from Luciole-Studio/Misaka-Agent.

Luciole-Studio/Misaka-Agent1251 repo~3.7kAutomated safety check: PassMITtoday
15
15.Tao Finetune Nv Tesseract ForecastingOfficial

NV-Tesseract Forecasting — transformer-based multivariate time series forecasting with DARR (context-enhanced kNN retrieval), interpretability, and fine-tuning.

NVIDIA/skills3.5k—~3.2kAutomated safety check: NotesApache-2.0today
16
16.Consent And Agency

Designing for informed user consent, opt-out, and human override.

Owl-Listener/ai-design-skills180—~598Automated safety check: PassMIT3 mo ago
17
17.SDK Implementation Guide

Operational guide for implementing a new Envilder runtime SDK.

macalbert/envilder138—~1.9kAutomated safety check: PassMIT2 days ago
18
18.Nan Safe Correlation

Per-feature NaN-safe Spearman/Pearson correlation across many features (genes, proteins, variants) with missing values.

jaechang-hits/SciAgent-Skills3701 repo~2.9kAutomated safety check: PassCC-BY-4.08 days ago
19
19.Endpoint Definition Designer

Designs primary, secondary, and exploratory endpoints for biomedical and clinical research protocols.

aipoch/medical-research-skills2k—~4.3kAutomated safety check: PassMIT20 days ago
20
20.Clinical Cohort Protocol Designer

Designs retrospective or prospective clinical cohort study protocols for biomedical and clinical research.

aipoch/medical-research-skills2k—~5.7kAutomated safety check: PassMIT20 days ago
21
21.AI Super Intelligence

Your AI research and engineering brain trust. An agent skill from majiayu000/claude-skill-registry.

majiayu000/claude-skill-registry6661 repo~3.7kAutomated safety check: PassMITtoday
22
22.Aaai Topic Selection

A skill your agent uses when deciding whether a project is a strong AAAI submission across its broad AI scope, should be reframed or routed to a dedicated track such as AI for Social Impact or AI…

brycewang-stanford/Awesome-Journal-Skills1.2k—~1.4kAutomated safety check: PassMIT10 days ago
23
23.Bioconductor Escher

The creation of effective visualizations is a fundamental component of data analysis.

bioMate-AI/biomate-bioconductor-kb804—~1.7kAutomated safety check: PassUnknown3 mo ago

Questions, answered from the data.

What is the best AI interpretability skill?

Esmfold2 from JimLiu/science-skills ranks first of the 23 AI interpretability skills listed here, with the highest score: its repository has 227 GitHub stars, 4 other GitHub owners carry a copy, its SKILL.md loads about 2.5k tokens and it passes the automated safety check with no findings. Next come Obliteratus and Anomalib Adding A Model.

Which AI interpretability skills are official?

1 of the 23 AI interpretability skills are official, published by the vendor's own GitHub organization: Tao Finetune Nv Tesseract Forecasting.

How are these skills ranked?

By Skill Navigator score, which combines the GitHub stars of the skill's repository (shared across that repo's skills and discounted for large collections), how many other GitHub owners carry a copy of the skill, and automated SKILL.md quality checks, minus penalties for safety-check warnings and for each further skill from the same repository. Skills that fail the safety check are not listed.

Category

More topics in AI & LLM Engineering

Products these skills work with

Roles that use these skills