context-doctor
Context and token optimizer for LLM agents: see where Claude Code, Cursor, Codex and Gemini CLI tokens go, then cut them: an every-prompt hook, smarter compaction, an autopilot proxy that edits history only when the prompt cache is cold, RAG chunk packing
as observed 2026-10-18T21:45:32.638Z- Identifier
context-doctor- Source
- npm
- Version observed
- 0.34.0
- Source repository
- https://github.com/KushalP1/context-doctor
- Repository observation
- Commits observed in the last 90 days
- First observed here
- 2026-09-02T06:49:03.117Z
- Observations recorded
- 17
- Weekly downloads (upstream)
- 1,187
- Publish rights held by
- kushalprakash
- Declared license
- MIT
Observation history
2026-10-18T21:45:32.638Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.33.0" | "0.34.0" |
2026-10-17T20:48:50.166Z
Fields that differed: description keywords version
| Field | Before | After |
|---|---|---|
description |
"Context and token optimizer for LLM agents: see where Claude Code, Cursor, Codex and Gemini CLI tokens go, then cut them: an every-prompt hook, smarter compaction, a cache-safe pr | "Context and token optimizer for LLM agents: see where Claude Code, Cursor, Codex and Gemini CLI tokens go, then cut them: an every-prompt hook, smarter compaction, an autopilot pr |
keywords |
["agent","ai-agents","ai-tools","anthropic","anthropic-sdk","autocompact","chunking","claude","claude-code","claude-desktop","codex","compaction","context-compression","context-eng | ["agent","ai-agents","ai-tools","anthropic","anthropic-sdk","autocompact","chunking","claude","claude-code","claude-desktop","codex","compaction","context-compression","context-eng |
version |
"0.32.0" | "0.33.0" |
2026-10-14T17:39:08.108Z
Fields that differed: description keywords version
| Field | Before | After |
|---|---|---|
description |
"Cut your Claude Code bill, measured on your own sessions: `npx context-doctor savings` shows what each lever saves, then the hook, autopilot and an earlier auto-compact window do | "Context and token optimizer for LLM agents: see where Claude Code, Cursor, Codex and Gemini CLI tokens go, then cut them: an every-prompt hook, smarter compaction, a cache-safe pr |
keywords |
["agent","ai-agents","anthropic","autocompact","claude","claude-code","claude-desktop","codex","context-engineering","context-management","context-window","cost","cursor","develope | ["agent","ai-agents","ai-tools","anthropic","anthropic-sdk","autocompact","chunking","claude","claude-code","claude-desktop","codex","compaction","context-compression","context-eng |
version |
"0.31.0" | "0.32.0" |
2026-10-12T15:36:43.577Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.29.0" | "0.31.0" |
2026-10-10T14:15:52.569Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.28.0" | "0.29.0" |
2026-10-09T13:20:37.397Z
Fields that differed: description version
| Field | Before | After |
|---|---|---|
description |
"See what your Claude Code sessions cost and cut it: offers /compact when you return to a large session after the cache expired (every surface, desktop app included), and a cache-s | "Cut your Claude Code bill, measured on your own sessions: `npx context-doctor savings` shows what each lever saves, then the hook, autopilot and an earlier auto-compact window do |
version |
"0.26.0" | "0.28.0" |
2026-10-07T11:52:53.973Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.24.0" | "0.26.0" |
2026-10-03T08:46:10.282Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.23.1" | "0.24.0" |
2026-10-01T06:47:50.339Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.22.0" | "0.23.1" |
2026-09-30T05:46:36.215Z
Fields that differed: description publisher version
| Field | Before | After |
|---|---|---|
description |
"Keep AI coding sessions lean automatically: clears stale tool output from every Claude Code and GPT API request, only when it costs nothing extra (measured 9.8% less input, no ses | "See what your Claude Code sessions cost and cut it: offers /compact when you return to a large session after the cache expired (every surface, desktop app included), and a cache-s |
publisher |
"kushalprakash" | "GitHub Actions" |
version |
"0.21.0" | "0.22.0" |
2026-09-29T04:47:03.725Z
Fields that differed: description keywords version
| Field | Before | After |
|---|---|---|
description |
"Profile and optimize LLM context windows. See what's eating your tokens and fix it — works with Claude Code, Claude Desktop, Cursor, Codex (OpenAI), and any MCP-capable AI app." | "Keep AI coding sessions lean automatically: clears stale tool output from every Claude Code and GPT API request, only when it costs nothing extra (measured 9.8% less input, no ses |
keywords |
["claude","claude-code","codex","context-management","context-window","cursor","llm","mcp","model-context-protocol","openai","prompt-caching","prompt-engineering","tokens"] | ["agent","ai-agents","anthropic","autocompact","claude","claude-code","claude-desktop","codex","context-engineering","context-management","context-window","cost","cursor","develope |
version |
"0.18.0" | "0.21.0" |
2026-09-24T00:36:14.200Z
Fields that differed: description keywords version
| Field | Before | After |
|---|---|---|
description |
"Profile and optimize LLM context windows. See what's eating your tokens and fix it — works with Claude, GPT, Gemini, and any MCP-capable AI app." | "Profile and optimize LLM context windows. See what's eating your tokens and fix it — works with Claude Code, Claude Desktop, Cursor, Codex (OpenAI), and any MCP-capable AI app." |
keywords |
["claude","context-management","context-window","llm","mcp","model-context-protocol","openai","prompt-caching","prompt-engineering","tokens"] | ["claude","claude-code","codex","context-management","context-window","cursor","llm","mcp","model-context-protocol","openai","prompt-caching","prompt-engineering","tokens"] |
version |
"0.14.2" | "0.18.0" |
2026-09-18T20:20:38.953Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.14.1" | "0.14.2" |
2026-09-17T19:21:47.151Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.13.5" | "0.14.1" |
2026-09-16T18:18:14.423Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.13.2" | "0.13.5" |
2026-09-09T11:17:27.221Z
Fields that differed: version
| Field | Before | After |
|---|---|---|
version |
"0.10.0" | "0.13.2" |
Correction
If you maintain this extension and believe anything above is inaccurate, request a correction. Corrections are published, and disputed entries are marked as disputed while under review.