Agent Native (Parallel Worker Dispatch)
A rulebook for delegating independent, non-overlapping tasks to parallel agent workers or selected persistent role sessions.
AutomationAdvanced★ 445⑂ 41AI score 6/10Last updated: Sep 24, 2026
What it does
- Defines the contract for dispatching independent tasks to parallel workers / role sessions.
- Requires validating every packet before the first launch: resolve symlinks, normalize paths, and compare write scopes case-insensitively so aliases can't hide collisions.
- Separates authority roles (Orchestrator, Implementer, Validator, Scribe) from cheap bounded delegations (Reader, Writer).
- Names the prompt-send optimism failure mode — a delivered prompt proves transport, not engagement — and the anti-pattern of restarting an unresponsive worker before capturing its state.
- Records runtime/session/context identity, provider state, transcript references, artifacts, and terminal status, returning them as runtime evidence rather than verdicts.
Who it's for
- Users actively running the AgentOps framework (boshu2/agentops) with executor adapters like NTM, Codex exec, or Agent Mail.
- Engineers designing multi-agent orchestration pipelines who want failure modes codified.
- Note: without familiarity with AgentOps vocabulary (packets, verdict.v2, write_scope), the barrier to entry is steep.
Example uses
- Three workers editing three modules: verify disjoint write scopes first, then dispatch each packet exactly once.
- A worker goes quiet for 10 minutes: capture observable state first instead of restarting and destroying the stall evidence.
- Reading a huge file without polluting the parent context: delegate to a cheap Reader with ≤350-line slices and get line-referenced bullets back.
· · · 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)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the 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.
- Open a terminal.
- Clone the repository:
git clone https://github.com/boshu2/agentops.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r agentops/images/gemini/skills/agent-native ~/.claude/skills/ - Verify the
references/files (model-dispatch.md, context-budget-delegation.md, judgment-receipts.md) came along. Some references live in a sibling skill (cass) or inscripts/swarm/, so keeping the full clone around is safer. - Restart Claude Code and test with a request like "delegate these independent tasks to parallel workers" to confirm the skill triggers.
View source on GitHub ↗License: Apache-2.0