ContextOptimizer
Optimizes CLAUDE.md, AGENTS.md, and other agent context files in a repository using research-backed principles.
Skill by RooseveltAdvisors
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/rooseveltadvisors-contextoptimizer
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: security, research, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/contextoptimizer
What's inside
- Skill name
ContextOptimizer- Description
- Optimizes CLAUDE.md, AGENTS.md, and other agent context files in a repository using research-backed principles. USE WHEN optimize context files OR audit CLAUDE.md OR improve agents.md OR reduce context bloat OR review agent instructions. Based on "Evaluating AGENTS.md" (Gloaguen et al., 2025) findings.
- License
- MIT
- Instructions
- 46 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- name: only lowercase letters, digits and single hyphens allowed
Files (6)
- ResearchFindings.md
- SKILL.md
- .gitkeep · Tools/
- Audit.md · Workflows/
- Generate.md · Workflows/
- Optimize.md · Workflows/
SKILL.md (first lines)
# ContextOptimizer Analyzes and optimizes repository-level agent context files (CLAUDE.md, AGENTS.md, .cursorrules, etc.) using evidence-based principles from the paper "Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?" (Gloaguen et al., ICML 2025). ## Key Research Findings (Paper Reference) The paper found that context files **tend to reduce task success rates** while **increasing inference cost by 20%+**. The core problem: unnecessary requirements make tasks harder. However, **developer-written minimal context files slightly outperform having no context at all**. The skill applies these findings to produce optimized, minimal context files. ## Workflow Routing | Workflow | Trigger | File | |----------|---------|------| | **Audit** | "audit context files", "analyze CLAUDE.md" | `Workflows/Audit.md` | | **Optimize** | "optimize context files", "improve CLAUDE.md" | `Workflows/Optimize.md` | | **Generate** | "generate context file", "create CLAUDE.md from scratch" | `Workflows/Generate.md` | ## Examples **Example 1: Audit existing context files** ``` User: "Audit the context files in this repo" → Invokes Audit workflow → Scans for CLAUDE.md, AGENTS.md, .cursorrules, .github/copilot-instructions.md → Scores each file against research-backed anti-patterns → Returns report with specific line-level recommendations ``` **Example 2: Optimize a bloated CLAUDE.md** ``` User: "Optimize my CLAUDE.md" → Invokes Optimize workflow → Reads current file, detects anti-patterns → Produces optimized version: removes redundancy, trims codebase overviews, keeps only actionable constraints → Shows before/after diff for user approval ``` **Example 3: Generate minimal context from scratch** ``` User: "Generate a CLAUDE.md for this repo" → Invokes Generate workflow → Explores repo structure, build system, test commands → Produces minimal context file with only non-obvious, actionable requirements → Avoids codebase overviews (paper found them ineffective) ```
From RooseveltAdvisors/claude-agent-stack, MIT.
More from RooseveltAdvisors/claude-agent-stack
- 0.433score
AgentBrowser
RooseveltAdvisors
Automates browser interactions for web testing, form filling, screenshots, and data extraction.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Agents
RooseveltAdvisors
Dynamic agent composition and management system.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Art
RooseveltAdvisors
Visual content generation with Excalidraw hand-drawn aesthetic.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
BlogWriter
RooseveltAdvisors
Write and publish blog posts to yourdomain.com.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Browser
RooseveltAdvisors
Code-first Playwright automation via TypeScript scripts.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
BugBot
RooseveltAdvisors
Adversarial iterative code review that loops until clean.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
ContextArchitect
RooseveltAdvisors
Audit and optimize agent context files across Claude Code and OpenAI Codex.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
CreateSkill
RooseveltAdvisors
Create and validate skills.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
DevFlow
RooseveltAdvisors
CI/CD workflow enforcer that detects current dev stage and guides to the next step.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
CodeReview
RooseveltAdvisors
Code review and anti-pattern detection.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Prompting
RooseveltAdvisors
Meta-prompting system for dynamic prompt generation using templates, standards, and patterns.
SkillDFreshness: D★ 1Claude Code +9 - 0.423score
Scraping
RooseveltAdvisors
Web scraping via progressive escalation (Bright Data proxy) and social media platform actors (Apify).
SkillDFreshness: D★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Mar 1, 2026 (8 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from RooseveltAdvisors
- 0.433score
AgentBrowser
RooseveltAdvisors
Automates browser interactions for web testing, form filling, screenshots, and data extraction.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Art
RooseveltAdvisors
Visual content generation with Excalidraw hand-drawn aesthetic.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
BlogWriter
RooseveltAdvisors
Write and publish blog posts to yourdomain.com.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Browser
RooseveltAdvisors
Code-first Playwright automation via TypeScript scripts.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
BugBot
RooseveltAdvisors
Adversarial iterative code review that loops until clean.
SkillDFreshness: D★ 1Claude Code +9 - 0.428score
Agents
RooseveltAdvisors
Dynamic agent composition and management system.
SkillDFreshness: D★ 1Claude Code +9
Cite this page
ContextOptimizer skill. Agent Mart. https://agentmart.dev/skills/rooseveltadvisors-contextoptimizer (last updated 2026-10-24)