AgentSkillsCN

junjo-deployment-sync

调研上游 Junjo AI Studio 的最新版本，并将相应更新应用至 junjo-ai-studio-deployment-example 仓库。当被要求将此仓库与 mdrideout/junjo-ai-studio 或 mdrideout/junjo-ai-studio-minimal-build 同步时，可执行以下操作：更新发布标签，修改 docker-compose/.env/设置文档，或在保留部署示例约束（预构建的 Junjo 镜像、无 JUNJO_BUILD_TARGET，以及内置 Caddy 与 junjo-app 服务）的前提下，验证迁移就绪性。

SKILL.md
--- frontmatter
name: junjo-deployment-sync
description: Investigate upstream Junjo AI Studio releases and apply aligned updates to the junjo-ai-studio-deployment-example repository. Use when asked to sync this repo with mdrideout/junjo-ai-studio or mdrideout/junjo-ai-studio-minimal-build, bump release tags, update docker-compose/.env/setup docs, or validate migration readiness while preserving deployment-example constraints (prebuilt Junjo images, no JUNJO_BUILD_TARGET, and included Caddy plus junjo-app services).

Junjo Deployment Sync

Use this skill to perform repeatable release-sync updates for this deployment example repository.

Workflow

  1. •Run the release audit helper:
    • •./.agents/skills/junjo-deployment-sync/scripts/release_audit.sh <target-tag>
    • •If <target-tag> is omitted, the script resolves the latest upstream release tag.
  2. •Read references/repo-constraints.md before editing.
  3. •Compare upstream and minimal-build deltas for these files first:
    • •docker-compose.yml
    • •.env.example
    • •scripts/junjo
    • •README.md
  4. •Apply updates in this repository while preserving deployment-example-specific behavior.
  5. •Run validations and report outcomes, including any intentionally preserved differences.

Update Rules

Compose and Runtime

  • •Bump Junjo core images to the target release in docker-compose.yml.
  • •Keep caddy and junjo-app services in place.
  • •Keep this repository in prebuilt-image mode for Junjo core services.
  • •Do not introduce JUNJO_BUILD_TARGET.
  • •Wire backend memory/DataFusion passthrough env vars if present upstream/minimal-build.

Environment Template

  • •Ensure .env.example exposes required tuning variables used by compose.
  • •Keep production URL placeholders compatible with setup automation.
  • •Keep Caddy/Cloudflare variables documented for this deployment example.

Setup Script

  • •Keep scripts/junjo focused on .env generation/update.
  • •Support production hostname-derived URLs.
  • •Support Cloudflare token setup for production (CLOUDFLARE_API_TOKEN).
  • •Preserve safe backup behavior (<env-file>.bak only when target exists).

Documentation

  • •Keep Quick Start concise.
  • •Present script setup as convenience and retain manual setup steps.
  • •Explicitly state Cloudflare token requirement for production Caddy SSL/DNS challenge.

Validation

Run this validation sequence after edits:

bash
python3 -m py_compile scripts/junjo
./scripts/junjo setup --dry-run --non-interactive --env development
./scripts/junjo setup --dry-run --non-interactive --env production --hostname junjo.example.com --cloudflare-token test_token
docker compose --env-file .env.example config

Also verify:

  • •No unintended introduction of JUNJO_BUILD_TARGET in this repo.
  • •docker-compose.yml still includes both caddy and junjo-app.
  • •README and script behavior stay consistent.

Resources

  • •scripts/release_audit.sh: Pull upstream references and print release-oriented deltas.
  • •references/repo-constraints.md: Hard constraints and file-specific guardrails for this repository.