AugmentClaude
EducationUI/UXML

Alibi Explainer

By a5c-ai· a5c-ai/babysitter· 0

Generate counterfactual explanations and trust scores to understand AI model decisions.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.
    Install
    Sign up to copy
    Sign up free to reveal

    This copies the whole skill folder into ~/.claude/skills/alibi-explainer-a5c-ai/ — the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
    Sign up free to reveal
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Alibi explainability skill for counterfactual explanations, anchors, and trust scores.

What this skill does

alibi-explainer

Overview

Alibi explainability skill for counterfactual explanations, anchors, trust scores, and advanced model interpretation techniques.

Capabilities

  • Counterfactual instance generation
  • Anchor explanations (rule-based)
  • Integrated gradients for deep learning
  • Kernel SHAP integration
  • Contrastive Explanation Method (CEM)
  • Trust scores for prediction confidence
  • Pertinent positives and negatives
  • Prototype and criticism selection

Target Processes

  • Model Interpretability and Explainability Analysis
  • Model Evaluation and Validation Framework

Tools and Libraries

  • Alibi
  • Alibi Detect
  • TensorFlow/PyTorch
  • scikit-learn

Input Schema

{
  "type": "object",
  "required": ["modelPath", "explainerType", "instancePath"],
  "properties": {
    "modelPath": {
      "type": "string",
      "description": "Path to the trained model"
    },
    "explainerType": {
      "type": "string",
      "enum": ["counterfactual", "anchor", "integrated_gradients", "cem", "trust_score", "prototype"],
      "description": "Type of Alibi explainer to use"
    },
    "instancePath": {
      "type": "string",
      "description": "Path to instance(s) to explain"
    },
    "counterfactualConfig": {
      "type": "object",
      "properties": {
        "targetClass": { "type": "integer" },
        "maxIterations": { "type": "integer" },
        "lambda": { "type": "number" },
        "featureRange": { "type": "object" }
      }
    },
    "anchorConfig": {
      "type": "object",
      "properties": {
        "threshold": { "type": "number" },
        "coverageSamples": { "type": "integer" },
        "beamSize": { "type": "integer" }
      }
    },
    "cemConfig": {
      "type": "object",
      "properties": {
        "mode": { "type": "string", "enum": ["PP", "PN"] },
        "kappaMin": { "type": "number" },
        "kappaMax": { "type": "number" }
      }
    },
    "trainingDataPath": {
      "type": "string",
      "description": "Path to training data (required for some explainers)"
    }
  }
}

Output Schema

{
  "type": "object",
  "required": ["status", "explanations"],
  "properties": {
    "status": {
      "type": "string",
      "enum": ["success", "error"]
    },
    "explanations": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "instanceId": { "type": "string" },
          "originalPrediction": { "type": "string" },
          "explanation": { "type": "object" }
        }
      }
    },
    "counterfactuals": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "instanceId": { "type": "string" },
          "counterfactual": { "type": "object" },
          "targetClass": { "type": "string" },
          "changedFeatures": { "type": "array" }
        }
      }
    },
    "anchors": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "instanceId": { "type": "string" },
          "rules": { "type": "array", "items": { "type": "string" } },
          "precision": { "type": "number" },
          "coverage": { "type": "number" }
        }
      }
    },
    "trustScores": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "instanceId": { "type": "string" },
          "score": { "type": "number" },
          "closestClass": { "type": "string" }
        }
      }
    }
  }
}

Usage Example

{
  kind: 'skill',
  title: 'Generate counterfactual explanations',
  skill: {
    name: 'alibi-explainer',
    context: {
      modelPath: 'models/loan_classifier.pkl',
      explainerType: 'counterfactual',
      instancePath: 'data/rejected_applications.csv',
      counterfactualConfig: {
        targetClass: 1,
        maxIterations: 1000,
        lambda: 0.1
      },
      trainingDataPath: 'data/train.csv'
    }
  }
}

Related skills

L

Learn Mistake

ilang-ai

Record and check past mistakes to avoid repeating them in similar tasks.

P

Pattern Learner

ilang-ai

Detect recurring patterns in your project and apply them automatically next time.

L

Learn Preference

ilang-ai

Remember your coding style and apply it automatically to future work.

P

Persistent Memory

a5c-ai

Store and retrieve decisions, discoveries, and patterns across chat sessions.