Agent skill

Context Mode Routing Rules

by mksglu in mksglu/context-mode

Rules for sending data-heavy work through context-mode's sandboxed MCP tools so raw output, web pages and large files stay out of the context window.

Custom licenceAuto-check passedAgent Workflows

Install Context Mode Routing Rules

skills CLI
$ npx skills add mksglu/context-mode --skill context-mode -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install mksglu/context-mode context-mode --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/mksglu/context-mode.git skills-src && mkdir -p .claude/skills && cp -r skills-src/configs/copilot-cli/skills/context-mode .claude/skills/context-mode && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
context-mode
GitHub stars
26k
Token cost
~425 tokens
SKILL.md length
175 words
Files
1
Skills in repo
11
Repo updated
First seen
Licence
Custom licence

At a glance

Rules for sending data-heavy work through context-mode's sandboxed MCP tools so raw output, web pages and large files stay out of the context window.

  • Works in 4 steps: GATHER: ctx_batch_execute(commands,… → PROCESS: ctx_execute / ctx_execute_file… → WEB: ctx_fetch_and_index(url, source)… → …
  • Analyzing a large log, CSV or API response without flooding the context window
  • SKILL.md covers Think in Code — MANDATORY, BLOCKED — do NOT use and Tool selection
  • Calls node and python

What it does

This short rule set is meant for GitHub Copilot CLI with the context-mode MCP tools installed. Its main principle is to write code for any analysis instead of reading data: to count, filter, compare, search, parse or transform, the agent runs a script through ctx_execute and prints only the answer, rather than loading raw data into the conversation.

It blocks curl and wget, inline HTTP calls made from node or python, and reading large files just to analyze them. In their place it names ctx_fetch_and_index followed by ctx_search for web content, ctx_execute_file for files, ctx_batch_execute to run several commands and search them in one call, and ctx_search with all questions batched. Artifacts are written to files and reported as a path with a one-line description.

When your agent uses it

  • Analyzing a large log, CSV or API response without flooding the context window
  • Fetching and searching a web page while keeping raw HTML out of the conversation
  • Running several data-heavy shell commands in one batched call

Example prompts

  • “Count the distinct error codes in logs/app.log and show me only the totals.”
  • “Fetch the release notes page and answer which version dropped Node 18 support.”
  • “Compare the two JSON exports in exports/ and list only the differing keys.”

Requirements

  • The context-mode MCP server and its ctx tools
  • GitHub Copilot CLI

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. GATHER: ctx_batch_execute(commands, queries) — runs all commands,
  2. PROCESS: ctx_execute / ctx_execute_file — sandbox; only stdout enters
  3. WEB: ctx_fetch_and_index(url, source) then ctx_search(queries) — raw
  4. SEARCH: ctx_search(queries: [...]) — all questions in one call.

What it can do on your machine

Read from SKILL.md and the folder at commit 2a7a92a. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • node
    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Context Mode Routing Rules loads about 425 tokens when it runs. Until then it costs about 57 tokens; SKILL.md has 175 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~57
When it runs · the whole SKILL.md, loaded when a task matches
~425

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 175 words (~425 tokens).

“context-mode MCP tools are available in GitHub Copilot CLI. These rules protect the context window from flooding — one unrouted command can dump tens of KB into the conversation. Follow them strictly.”

— opening of SKILL.md by mksglu, Custom licence
name
context-mode

Read the full SKILL.md on GitHub

Files

Just SKILL.md in configs/copilot-cli/skills/context-mode of mksglu/context-mode.

Open the folder on GitHubat commit 2a7a92a

Compare with similar skills

Context Mode Routing Rules next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Context Mode Routing Rules compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Context Mode Routing Rules this skillmksglu/context-mode26k—~425Automated safety check: PassCustom licence
LemmalogJordyZomer/lemmalog329—~2.8kAutomated safety check: PassMIT
Cortex Mem MCPsopaco/cortex-mem312—~2.8kAutomated safety check: PassMIT
MCP Code Search Tool SelectionContext-Engine-AI/Context-Engine402—~1.3kAutomated safety check: PassMIT
Organizemcpware/cross-code-organizer382—~246Automated safety check: NotesMIT
Entroly Context Controljuyterman1000/entroly473—~501Automated safety check: PassApache-2.0

Similar skills

  • Lemmalog

    JordyZomer/lemmalog

    Externalize working memory and logical state into the lemmalog Datalog engine (MCP).

    329 GitHub stars~2.8k tokensUpdated 7 days ago
    Agent WorkflowsAuto-check passed
  • Cortex Mem MCP

    sopaco/cortex-mem

    Persistent memory enhancement for AI agents. An agent skill from sopaco/cortex-mem.

    312 GitHub stars~2.8k tokensUpdated 2 mo ago
    Agent WorkflowsAuto-check passed
  • MCP Code Search Tool Selection

    Context-Engine-AI/Context-Engine

    Rules for choosing Qdrant-Indexer semantic search over grep or file reads when exploring code, debugging or asking where and why questions.

    402 GitHub stars~1.3k tokensUpdated 3 mo ago
    Agent WorkflowsAuto-check passed
  • Organize

    mcpware/cross-code-organizer

    Open the Cross-Code Organizer (CCO) dashboard — view and manage all memories, skills, MCP servers, hooks, and configs across scopes

    382 GitHub stars~246 tokensUpdated 2 days ago
    Agent WorkflowsAuto-check: notes
  • Entroly Context Control

    juyterman1000/entroly

    Surgically select, compress, and recover codebase context using Entroly's MCP tools.

    473 GitHub stars~501 tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Claude Code Mastery Squad

    ohmyjahh/xquads-squads

    Routes a request to one of eight specialist agents covering hooks, skills, subagents, MCP integration and context engineering for Claude Code.

    276 GitHub stars~1.1k tokensUpdated 9 days ago
    Agent WorkflowsAuto-check passed

More from mksglu/context-mode

All 11 skills in this repo
  • Context Mode Output Sandbox

    mksglu/context-mode

    Routes large command, file, API and browser output through context-mode tools so only the needed result enters the agent's context, instead of dumping it via Bash.

    26k GitHub stars~4.1k tokensUpdated yesterday
    Auto-check passed
  • Routing rules for using context-mode MCP tools in Antigravity CLI: sandboxed code runs, file analysis, indexed search and web fetches that keep large output out of the conversation.

    26k GitHub stars~850 tokensUpdated yesterday
    Auto-check passed
  • Context Mode Ops

    mksglu/context-mode

    Runs maintenance for the context-mode project with parallel subagents: issue triage, PR review, releases, bug fixes, announcements and branch syncing.

    26k GitHub stars~6k tokensUpdated yesterday
    Auto-check passed
  • Context Mode Indexer

    mksglu/context-mode

    Indexes a local file or project directory into context-mode's FTS5 knowledge base so later searches return focused snippets instead of rereading whole files.

    26k GitHub stars~328 tokensUpdated yesterday
    Auto-check passed
  • Context Mode Knowledge Purge

    mksglu/context-mode

    Permanently deletes a project's or a single session's indexed context-mode knowledge base, after confirming the scope and warning what will be lost.

    26k GitHub stars~671 tokensUpdated yesterday
    Auto-check passed
  • Context Mode Doctor

    mksglu/context-mode

    Run context-mode diagnostics. Checks runtimes, hooks, FTS5, plugin registration, npm and marketplace versions. Trigger: /context-mode:ctx-doctor

    26k GitHub stars~262 tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Context Mode Routing Rules

What does Context Mode Routing Rules do?

Rules for sending data-heavy work through context-mode's sandboxed MCP tools so raw output, web pages and large files stay out of the context window. This short rule set is meant for GitHub Copilot CLI with the context-mode MCP tools installed. Its main principle is to write code for any analysis instead of reading data: to count, filter, compare, search, parse or transform, the agent runs a script through ctx_execute and prints only the answer, rather than loading raw data into the conversation.

When should I use Context Mode Routing Rules?

Context Mode Routing Rules fits situations like: analyzing a large log, CSV or API response without flooding the context window; fetching and searching a web page while keeping raw HTML out of the conversation; running several data-heavy shell commands in one batched call.

How do I install Context Mode Routing Rules in Claude Code?

Run `npx skills add mksglu/context-mode --skill context-mode -a claude-code`. Or copy the skill folder (configs/copilot-cli/skills/context-mode in mksglu/context-mode) into .claude/skills/context-mode in your project. Claude Code loads it when a task matches its description.

How do I install Context Mode Routing Rules in Codex?

Run `npx skills add mksglu/context-mode --skill context-mode -a codex`. Or copy the skill folder (configs/copilot-cli/skills/context-mode in mksglu/context-mode) into .agents/skills/context-mode in your project. Codex loads it when a task matches its description.

Can I use Context Mode Routing Rules in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add mksglu/context-mode --skill context-mode -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/context-mode, .gemini/skills/context-mode, .github/skills/context-mode and .opencode/skills/context-mode in your project.

What does Context Mode Routing Rules need to run?

Going by SKILL.md and its folder, Context Mode Routing Rules needs the command-line tools its instructions call (node and python). Our summary lists: The context-mode MCP server and its ctx tools; GitHub Copilot CLI.

Does Context Mode Routing Rules access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Context Mode Routing Rules safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Context Mode Routing Rules use?

Context Mode Routing Rules has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Context Mode Routing Rules use?

About 425 tokens (SKILL.md is roughly 1.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Context Mode Routing Rules?

Skills that share tags, products or a category with Context Mode Routing Rules: Lemmalog (JordyZomer/lemmalog, 329 stars), Cortex Mem MCP (sopaco/cortex-mem, 312 stars), MCP Code Search Tool Selection (Context-Engine-AI/Context-Engine, 402 stars) and Organize (mcpware/cross-code-organizer, 382 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Context Mode Routing Rules?

mksglu (a GitHub user) maintains it in mksglu/context-mode, which has 25,647 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 7, 2026.

Source: mksglu/context-mode on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.