Gain — Development skill for Claude Code
Show how many context tokens mcp-multiplexer is saving you (startup withholding + on-demand schema serving).
How to install Gain
Installs to ~/.claude/skills/johgirard-mcp-multiplexer-gain/SKILL.md
mkdir -p ~/.claude/skills/johgirard-mcp-multiplexer-gain && curl -fsSL https://raw.githubusercontent.com/JohGirard/mcp-multiplexer/HEAD/commands/gain.md -o ~/.claude/skills/johgirard-mcp-multiplexer-gain/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Gain does
description: Show how many context tokens mcp-multiplexer is saving you (startup withholding + on-demand schema serving)
Run:
mcp-multiplexer --stats
If the binary is not found, tell the user to install it first (see the setup skill) — do not guess numbers.
Present the output compactly in chat:
- Startup saved is the headline: tokens that would have been injected by connecting all upstream servers directly, minus what the mux's own meta-tools cost.
- **On-demand sc
Alternatives in Development
- Openwolf — Sharper context 699 ★
- Mcptoon — Listing 255 MCP tools costs 71,929 tokens 195 ★
- Seance Skill — Seance Terminal Multiplexer 130 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Openclaw Nim Skill
OpenClaw skill for offloading heavy tasks to NVIDIA NIM models and saving context tokens
Continuous Improvement
Reflect on the current session, analyze observations for patterns, and show instinct status. Runs on-demand to
Thumbgate Blocked
Show what ThumbGate has actually blocked — gate enforcement stats and the full enforcement matrix. Use for "wh
Warden Power
Zero-token power planner — from the agent's own recorded run-to-run variance, report the minimum detectable sa
Vllm Turing
Turing (sm75) speedups for vLLM: Qwen3.8-27B W8A8 on 2x RTX 2080 Ti with custom CUDA kernels, long-context ver
Crewcut Gain
Show the reading scoreboard — what Crewcut's discipline saved this session
Related Agents
Frame Describer
Describes video frames as detailed text. Used when frame_mode is "descriptions" to convert visual frames into
Raw Retriever
Context-isolated searcher of the owner's raw ledgers (KakaoTalk, Mattermost, Discord, Notion, Gmail) through t
Phase 1 Figma Harvest
Figma-input variant of Phase 1 in the Next.js clone pipeline. For one section, reads design data via the offic