Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Agent Native (role-shaped agent runtime)

Runs caller-selected agent sessions as four explicit roles — orchestrator, implementer, validator, scribe — with strict authority and evidence rules.

AutomationAdvanced★ 428⑂ 40AI score 6/10Last updated: Aug 23, 2026

What it does

  • Splits agent workers into four explicit roles (Orchestrator / Implementer / Validator / Scribe), each with an authority boundary that is checkable from its packet.
  • Enforces a contract: no worker starts without an explicit packet, role, workspace, context identity, and evidence destination.
  • Names a common failure mode — prompt-send optimism — where a delivered prompt is mistaken for an engaged worker; only observable state counts as proof.
  • Provides an escalation ladder for stuck workers ranked by cost and reversibility: observe → nudge → replace worker → restart runtime.
  • Treats ntm, codex-exec, native processes, Agent Mail and similar tools as replaceable adapters, and forbids adapters from issuing binding verdicts.

Who it's for

  • Developers and tech leads running multiple concurrent agent sessions who need clean work division.
  • Anyone who wants hard guarantees about which worker may touch which files.
  • Teams already using (or evaluating) AgentOps-style packet/evidence/verdict workflows.

Examples

  1. Hand a refactor to an implementer worker, then spawn a fresh read-only validator session to judge only the candidate output.
  2. Give two concurrent workers disjoint subjects (different modules) so their writes can never collide.
  3. When a worker goes silent, capture provider state and transcript references first instead of restarting — preserving the evidence of why it stalled.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/boshu2/agentops/HEAD/images/gemini/skills/agent-native/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the images/gemini/skills/agent-native folder from the GitHub repo boshu2/agentops into my ~/.claude/skills/agent-native/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/boshu2/agentops.git && mkdir -p ~/.claude/skills && cp -r agentops/images/gemini/skills/agent-native ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repository: git clone https://github.com/boshu2/agentops.git
  3. Create the skills folder if needed: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r agentops/images/gemini/skills/agent-native ~/.claude/skills/
  5. Verify that references/model-dispatch.md came along; if missing, locate it in the repo and place it at the same relative path.
  6. Restart Claude Code and invoke it with phrases like "agent-native factory", "role-shaped agent panes", or "persistent workers".
  7. Adapters such as ntm or codex-exec are optional, but you must install and configure them separately for real multi-worker runs.
View source on GitHub ↗License: Apache-2.0