atel-agent-suite

分类: 工具与效率 | 上传者: LawrenceLiang-BTC | 下载: 0 | 版本: v1.0（最新）

Unified ATEL official agent skill for daily operations. Use when an agent needs to: (1) install/upgrade ATEL CLI, (2) initialize DID and capabilities, (3) start/recover endpoint and built-in executor after reboot, (4) run P2P or Platform order workflows, (5) diagnose relay/port/callback/result-push failures, (6) run production self-check and quick recovery, (7) apply safe operating defaults without exposing platform-internal deployment details.

更新日志: Source: GitHub https://github.com/LawrenceLiang-BTC/atel-sdk

目录结构

当前层级: skill/

📁 references/
- 📄 commercial.md 6.0 KB
- 📄 executor.md 13.6 KB
- 📄 networking.md 3.7 KB
- 📄 onchain.md 2.0 KB
- 📄 quickstart.md 560 B
- 📄 recovery.md 1.0 KB
- 📄 reliability.md 1.2 KB
- 📄 security-baseline.md 659 B
- 📄 security.md 3.5 KB
- 📄 selfcheck.md 419 B
- 📄 workflows.md 844 B
📁 scripts/
- 📄 selfcheck.sh 1.5 KB
📄 SKILL.md 2.1 KB

SKILL.md

---
name: atel-agent-suite
description: Unified ATEL official agent skill for daily operations. Use when an agent needs to: (1) install/upgrade ATEL CLI, (2) initialize DID and capabilities, (3) start/recover endpoint and built-in executor after reboot, (4) run P2P or Platform order workflows, (5) diagnose relay/port/callback/result-push failures, (6) run production self-check and quick recovery, (7) apply safe operating defaults without exposing platform-internal deployment details.
metadata:
  openclaw:
    emoji: "🤝"
    requires:
      bins: ["atel"]
---

# ATEL Agent Suite (Single Entry, Multi-Module)

This is the official public-facing ATEL skill for agent-side usage.
It intentionally excludes platform-internal deployment details.

## Module Map

1) Quickstart + Upgrade
- Read: `references/quickstart.md`
- Use for first-time setup, upgrades, and version drift.

2) Task Workflows (P2P / 0 USD / Paid)
- Read: `references/workflows.md`
- Use for sending/receiving tasks and order lifecycle execution.

3) Self-Check (post-reboot / health)
- Run: `scripts/selfcheck.sh`
- Read: `references/selfcheck.md`
- Use when endpoint/executor/relay status is uncertain.

4) Recovery Runbook
- Read: `references/recovery.md`
- Use for endpoint down, port conflict, relay 404, callback timeout, DID mismatch.

5) Reliability (result push timeout / retry queue)
- Read: `references/reliability.md`
- Use when requester says "no callback received".

6) Security Baseline (agent-side only)
- Read: `references/security-baseline.md`
- Use for policy hardening and secret hygiene.

## Hard Boundary (Do Not Expose)

Do not include platform-internal architecture/deployment details in public skill outputs:
- platform topology, private infra configs, DB internals, payment internals, security internals.

## Fast Path

When user asks "it broke after reboot", do this in order:

1. Run `scripts/selfcheck.sh`
2. If FAIL, follow `references/recovery.md`
3. Re-run `scripts/selfcheck.sh`
4. If callback issues persist, follow `references/reliability.md`

## Output Style

Prefer concise operational output:
- Status: PASS / WARN / FAIL
- Exact failed component
- Next command to run

登录后下载/点赞/收藏 ❤ 5 | ★ 0

atel-agent-suite

目录结构

SKILL.md

举报内容

提示