skills.rest
jadecli
jadecliOFFICIAL
Explore all skills by this author →
GitHub

context-audit

Audit conversation context usage and prescribe token compaction strategies.

1|1|Updated Mar 27, 2026
Search Tags:#context#optimization#audit#agent#compaction#token
One-click install
npx skills add https://github.com/jadecli/researchers --skill context-audit-jadecli
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-audit
Source: https://github.com/jadecli/researchers/tree/main/claude-multi-agent-sdk/.claude/skills/context-audit
Command: npx skills add https://github.com/jadecli/researchers --skill context-audit-jadecli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audit current context window usage and recommend compaction strategies to reduce token waste and latency in long-running sessions.

Core Features & Use Cases

  • Measure current context usage (loaded CLAUDE.md files, tool definitions, and history tokens).
  • Detect waste (excess history, verbose outputs) and propose targeted truncation or summarization.
  • Generate action plans aligned with internal strategy rules (e.g., prioritize recent decisions, clear unused tool results) for efficient context management.
  • Use Case: When sessions feel sluggish or approach context limits, apply compaction to maintain performance.

Quick Start

Run the context-audit skill on an active session to receive a compacting plan and recommended next steps.

Frequently Asked Questions about context-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reduce token usage in a long-running conversation session?▼

To reduce token usage in a long-running conversation, audit current context to measure loaded files and history tokens, then apply compaction strategies like summarizing history or truncating verbose tool outputs to stay within budget.

What is context compaction and when should I apply it?▼

Context compaction is the process of trimming, summarizing, or delegating conversation history to optimize token usage. Apply it during multi-turn dialogs or when the context window nears limits and session performance feels sluggish.

How do I estimate remaining token budget in an active agent session?▼

Estimate the remaining token budget by measuring loaded configuration files, tool definitions, and conversation history tokens. This audits current context usage against limits to determine how much space remains for new outputs.

What is the best way to clear unused tool results and optimize context?▼

The best way to optimize context is generating an action plan that prioritizes recent decisions and clears unused tool results. This targeted truncation aligns with internal manager rules to efficiently reduce token waste.

Are there limitations to automatic context summarization for multi-turn dialogs?▼

Limitations of context summarization include the potential loss of critical historical details when truncating. It prescribes actionable compaction strategies, but users must ensure recent decisions are prioritized to avoid losing important agent state.

Related Skills

Explore More Software Engineering Skills→
hotaqhotaq

help

Guide oh-my-codex setup, keyword routing, and usage analysis.

Community
Intermediate
abdullah1854abdullah1854
1

context-optimization

Optimize AI agent context usage by applying compaction, masking, caching, and partitioning techniques.

Community
Advanced
BrennonTWilliamsBrennonTWilliams
7

ll-compact-session

Compact and summarize session memory to reduce context footprint.

Community
Basic
korchasakorchasa
3

flow-skill-analyze-context

Estimate total token usage and costs for AI interactions.

Community
Intermediate