Context Pruner Skill

**Version:** 1.0.0

14 stars

Best use case

Context Pruner Skill is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

**Version:** 1.0.0

Teams using Context Pruner Skill should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/cfn-context-pruner/SKILL.md --create-dirs "https://raw.githubusercontent.com/masharratt/claude-flow-novice/main/.claude/cfn-extras/skills/advanced-features/cfn-context-pruner/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/cfn-context-pruner/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How Context Pruner Skill Compares

Feature / AgentContext Pruner SkillStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

**Version:** 1.0.0

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Context Pruner Skill

**Version:** 1.0.0
**Purpose:** Hierarchical context summarization for CFN Loop v3 iterations

## Overview

Reduces context size across iterations by:
- Keeping current iteration in full detail
- Summarizing previous iterations
- Extracting key themes from feedback
- Maintaining deliverable tracking

## Goal

Reduce context from 120 KB (iteration 10) to 15 KB (88% reduction)

## Usage

```bash
PRUNED_CONTEXT=$(./.claude/skills/context-pruner/prune-context.sh \
  --iteration 3 \
  --full-history "$FULL_HISTORY" \
  --current-context "$CURRENT_CONTEXT")

echo "$PRUNED_CONTEXT"
```

## Pruning Strategy

### Iteration 1
- Full detail (no pruning)
- Context size: ~5 KB

### Iteration 2
- Iteration 1: Summary only
- Iteration 2: Full detail
- Context size: ~8 KB

### Iteration 3+
- Iterations 1-(N-1): Summary with key themes
- Iteration N: Full detail
- Context size: ~10-15 KB (stable)

## Summary Format

```
Iterations 1-2 Summary:
- Initial confidence: 0.72
- Key feedback themes:
  * Add error handling
  * Improve test coverage
  * Address security concerns
- Final confidence: 0.82
- Progress: +0.10
```

## Key Themes Extraction

Analyze all feedback across iterations to extract recurring themes:

```bash
# Count feedback occurrences
"Add error handling" - 3 times (iterations 1, 2, 3)
"Improve test coverage" - 2 times (iterations 1, 2)
"Security concerns" - 2 times (iterations 2, 3)

# Output top 5 themes
```

## Integration

Used by:
- Main Chat loop orchestration
- Coordinator for iteration context building

Related Skills

cfn-mdap-context-injection

14
from masharratt/claude-flow-novice

Inject MDAP codebase files for troubleshooting (coordinator, implementers, validators, Docker, orchestration)

supabase-schema-sync

14
from masharratt/claude-flow-novice

Introspects Supabase DB after migrations and updates project db-query skill with current schema. Run after any migration to keep agent context accurate.

commit

14
from masharratt/claude-flow-novice

Stage, commit, and push changes using a background github-commit-agent. Accepts optional args for message override or push control.

cfn-vote-implement

14
from masharratt/claude-flow-novice

MUST BE USED after cfn-dry-review or cfn-alpha-launch:manifest produces a manifest. Also the verification phase of /cfn-loop-task. Do not manually implement code review suggestions - always route through this skill. 3-agent specialized voting. Unanimous (3/3) auto-implemented with TDD. 2/3 routed to product-owner agent. 1/3 surfaced to user via AskUserQuestion (batched 4 per call, at end).

cfn-utilities

14
from masharratt/claude-flow-novice

Reusable bash utility functions for CFN Loop - logging, error handling, retry, file operations. Use when you need structured logging, atomic file operations, retry logic with exponential backoff, or standardized error handling in bash scripts.

CFN Test Runner Skill

14
from masharratt/claude-flow-novice

**Version:** 1.0.0

cfn-test-framework

14
from masharratt/claude-flow-novice

Test execution, running, and webapp testing for CFN

cfn-task-planning

14
from masharratt/claude-flow-novice

Classify tasks, initialize structured configs with scope boundaries, decompose complex tasks

Specialist Injection Skill

14
from masharratt/claude-flow-novice

## Purpose

!/bin/bash

14
from masharratt/claude-flow-novice

# cfn-task-intelligence.sh

Task Complexity Estimator

14
from masharratt/claude-flow-novice

**Version:** 1.0.0

task-classifier

14
from masharratt/claude-flow-novice

Analyzes task descriptions and classifies them into categories for agent selection