smithery.ai

agent-protocol

Rules for efficient agent-human communication and high-performance development workflows.

First seen Mar 25, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,333 B
  • docs SUMMARY.md 111 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Skill: Agent Protocol

This skill defines a high-efficiency communication style and behavioral guidelines for AI agents, inspired by steipete/agent-scripts.

📡 Telegraph Style

  • Drop filler: Avoid polite preamble ("Sure!", "I can help with that") and redundant confirmation.
  • Telegraphic: Use concise, information-dense phrases.
  • Minimal Tokens: Optimize for context window efficiency and speed.

🛠️ Behavioral Guardrails

  • Root Cause First: Fix the source of a bug, not the symptoms. Don't apply "band-aids".
  • Observed Change: Only assume changes that are visible via git status or logs.
  • File Size Limit: Aim to keep files under 500 lines. Refactor and split when files grow too large.
  • Regression Tests: Always add a test case when fixing a bug to prevent regressions.

🔄 Workflow Patterns

  • Pointer-Style Rules: When working across multiple projects, reference a central AGENTS.MD if available to ensure rule consistency.
  • Commit Early: Use the committer script to create atomic, well-described commits.
  • Verify End-to-End: Prefer verifying the entire flow (backend + frontend) rather than isolated units when possible.