agentleFS
javimoschjavimosch/supercli/.agents/skills/context-mode/SKILL.md

context-mode

Skill53 starsChanged 5 months agoMITView on GitHub

Use context-mode (via supercli MCP) for large codebase analysis (50+ files), long command outputs, or repeated searches. Reduces token usage by up to 98%. NOT for simple commands or small file sets.

  • Installs packages

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    ctx_batch_execute: Default for exploration - runs commands, indexes, searches in one call

    line 46
  2. Distinctive:

    ctx_index + ctx_search: When you'll search the same content repeatedly

    line 47
  3. Distinctive:

    ctx_stats: Check token savings

    line 51

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

context-mode Decision Guide

context-mode is an MCP server accessed via supercli. It maintains a persistent knowledge base across sessions.

Quick Decision Tree

Task size?

  • Small (1-10 files, < 500 lines output) → Use direct bash/read_file
  • Large (50+ files, 1000+ lines output) → Use context-mode via supercli

Will you search the same content multiple times?

  • Yes → Use context-mode (index once, search many times)
  • No → Use direct tools

Most Common Workflow (via supercli)

# 1. Verify context-mode is available
sc mcp list --json

# 2. Verify installation (daemon auto-starts)
sc mcp call --mcp-server context-mode --tool ctx_doctor --input-json '{}' --json

# 3. Explore codebase
sc mcp call --mcp-server context-mode --tool ctx_batch_execute --input-json '{
  "commands": [
    {"label": "Structure", "command": "find . -type f -name \"*.js\" | head -100"},
    {"label": "Package", "command": "cat package.json"}
  ],
  "queries": ["entry point", "dependencies", "routes"]
}' --json

# 4. Search indexed content (if needed)
sc mcp call --mcp-server context-mode --tool ctx_search --input-json '{
  "queries": ["search terms"]
}' --json

Tool Selection Guide

  • ctx_batch_execute: Default for exploration - runs commands, indexes, searches in one call
  • ctx_index + ctx_search: When you'll search the same content repeatedly
  • ctx_execute: Single long-running command (npm test, build)
  • ctx_fetch_and_index: External documentation or API specs
  • ctx_purge: Clear knowledge base between unrelated tasks
  • ctx_stats: Check token savings

When NOT to Use

  • ls -la, cat package.json → Direct bash
  • Checking 1-10 files → read_file
  • Quick duplicate check → bash + uniq -c
  • Simple grep → Direct grep or Grep tool
  • Output < 500 lines → Direct tools

Installation (if not available)

npm install -g context-mode

Then add to your harness's MCP server configuration with command: context-mode

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.