What does Context Compactor actually do?

It distills large amounts of session data into concise summaries and logs, preventing the AI from becoming slow or confused by excessive conversation history.

Can I customize the context limits?

Yes, the skill respects a LINE_BUDGETS.yaml file where you can define specific limits for different types of context and logs.

How does it help with project handoffs?

It automatically generates a HANDOFF.md file that captures the current state, making it easy to resume work later or transfer context between different AI sessions.

Will I lose important information when it trims logs?

No, the skill includes guardrails to ensure that critical evidence and required project references are preserved while respecting defined line budgets.

Is this necessary for small projects?

While most effective for complex tasks, it provides value to any project by maintaining a clean backlog and clear action logs for better organization.

Context Compactor

Name: Context Compactor
Author: CheekyCodexConjurer

byCheekyCodexConjurer

Productivity & Workflow

Compresses extensive session history into structured handoff logs and backlogs to maintain peak AI performance.

About

The Context Compactor skill optimizes Claude's efficiency during long-running coding sessions by distilling bloated conversation history into manageable markdown files. By automatically generating handoff summaries, updating backlogs, and rotating context logs while adhering to strict line budgets, it ensures the model remains focused on the current task without losing sight of the project's overall state or critical references. This is essential for maintaining accuracy and speed in complex, multi-stage development workflows.

Key Features

Dynamic backlog updates for tracking open action items
Automated state summarization into HANDOFF.md files
0 GitHub stars
Configurable context limits via LINE_BUDGETS.yaml
Intelligent rotation and trimming of AUTO_CONTEXT.md
Compact action logging for historical auditability

Use Cases

Optimizing token usage and model focus during long-run development cycles
Managing large-scale refactors that span multiple AI sessions
Maintaining project continuity when switching between different coding tasks

About

Key Features

Dynamic backlog updates for tracking open action items
Automated state summarization into HANDOFF.md files
0 GitHub stars
Configurable context limits via LINE_BUDGETS.yaml
Intelligent rotation and trimming of AUTO_CONTEXT.md
Compact action logging for historical auditability

Use Cases

Optimizing token usage and model focus during long-run development cycles
Managing large-scale refactors that span multiple AI sessions
Maintaining project continuity when switching between different coding tasks