ワンクリックでManusで任意のスキルを実行

sync-plan-issues

スター3

フォーク0

更新日2026年4月11日 20:14

Use when the user wants to mirror chevp-ai-framework plan files (context/plans/**/*.md) to GitHub Issues, or says "sync plans", "issue für §X.Y", "plans als issues anlegen", "create issues from plans", "spiegle backlog nach github". Wraps integration/github-issues/sync-plans-to-issues.py.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

chevp

chevp/chevp-ai-framework

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

Sync Plan Files to GitHub Issues

Mirrors plan files in context/plans/<phase>/ to GitHub Issues, one issue per plan. Source-of-truth stays the markdown file; the issue body links back to it. Idempotent — re-running skips existing issues.

When to trigger

User asks to "sync plans", "create issues for backlog", "alle plans als github issues", "mirror plans to issues", or similar
User asks to create an issue for a specific plan (e.g. "issue für §5.2")
After a batch of new plans was added and the user wants the tracking board updated

Prerequisites

The project has a .plan-sync.json at its root (see ../../integration/github-issues/.plan-sync.example.json). If missing, stop and offer to create one based on the project's CLAUDE.md and existing context/plans/ layout. Do not guess repo or repo_url_prefix — ask.
gh is authenticated against the target repo (gh auth status).
Plan files use the framework template's frontmatter (../../templates/plan-template.md) — either id: (flat) or paragraph: (paragraph numbering), plus type: and ideally title:.

Steps

Verify config exists. If not, create one with the user (ask for repo slug; default plan_dirs to ["context/plans/backlog", "context/plans/active"]).
Always dry-run first:
```
python3 path/to/chevp-ai-framework/integration/github-issues/sync-plans-to-issues.py --dry-run --limit 3
```
Show the user 2–3 example issue titles + labels and confirm they look right before going wider.
Then run for the full set (without --limit) once the user approves the dry-run output.
For a single plan (e.g. "issue für §5.2"): use --limit 1 after temporarily reducing plan_dirs to the directory containing that plan, OR fall back to a direct gh issue create with the same title format the script would produce. Prefer the script for consistency.

Safety

This calls gh issue create and gh label create — both are user-visible writes to GitHub. Always confirm intent before the first non-dry-run invocation.
Never add --update without explicit user request — it overwrites existing issue bodies.
If the dry-run shows zero plans discovered, stop and investigate the plan_dirs config rather than assuming success.

Reference

Script + full docs: ../../integration/github-issues/README.md
Conceptual integration: ../../integration/kanban.md

このリポジトリの他の Skills

同じリポジトリ

create-prd-plan

chevp/chevp-ai-framework

Use when an approved EXP plan is ready to be implemented and a Production-Plan (PRD) is needed, or when the user says "PRD plan", "production plan", "produce", "implement EXP-NNN", "draft PRD". Produces a PRD per the chevp-ai-framework template.

2026-04-293

create-proposal

chevp/chevp-ai-framework

Use when an out-of-scope item, Challenger-identified failure mode, or NOT-in-Scope row warrants its own future plan (Rule 12). Trigger phrases: "create proposal", "PROP for this", "open a proposal", "log this for later", "this is out of scope but worth tracking". Produces a `PROP-NNN` file per the plan-proposal template.

2026-04-293

run-challenger

chevp/chevp-ai-framework

Use before requesting G2 approval for an EXP plan, or when scope changes are proposed mid-Production (Rule 11 / Challenger role). Trigger phrases: "run challenger", "challenge this plan", "stress-test the approach", "draft the challenger block", "what could go wrong with this plan". Produces the four mandatory Challenger sections per [02-exploration/challenger.md].

2026-04-293

write-insights

chevp/chevp-ai-framework

Use when an Exploration cycle is wrapping up and the `insights.md` learning artifact is needed (mandatory before G2), or when Production has revealed implementation surprises that must be appended (Rule 9 / Learning Loop). Trigger phrases: "write insights", "insights file", "what did we learn", "log insights", "update insights with implementation surprises".

2026-04-293

create-adr

chevp/chevp-ai-framework

Use when the user wants to record an architecture decision, says "write an ADR", "new ADR", "document this decision", or when a non-trivial architectural choice is made during Context or Exploration that future contributors need to understand.

2026-04-043

create-ctx-plan

chevp/chevp-ai-framework

Use when the user starts a new task and a Context-Plan (CTX) is needed, or when they say "context plan", "CTX", "start context step", or a new request arrives with no active plan. Produces a CTX plan per the chevp-ai-framework template.

2026-04-043

name	sync-plan-issues
description	Use when the user wants to mirror chevp-ai-framework plan files (context/plans/*/.md) to GitHub Issues, or says "sync plans", "issue für §X.Y", "plans als issues anlegen", "create issues from plans", "spiegle backlog nach github". Wraps integration/github-issues/sync-plans-to-issues.py.

Sync Plan Files to GitHub Issues

When to trigger

User asks to "sync plans", "create issues for backlog", "alle plans als github issues", "mirror plans to issues", or similar
User asks to create an issue for a specific plan (e.g. "issue für §5.2")
After a batch of new plans was added and the user wants the tracking board updated

Prerequisites

The project has a .plan-sync.json at its root (see ../../integration/github-issues/.plan-sync.example.json). If missing, stop and offer to create one based on the project's CLAUDE.md and existing context/plans/ layout. Do not guess repo or repo_url_prefix — ask.
gh is authenticated against the target repo (gh auth status).
Plan files use the framework template's frontmatter (../../templates/plan-template.md) — either id: (flat) or paragraph: (paragraph numbering), plus type: and ideally title:.

Steps

Verify config exists. If not, create one with the user (ask for repo slug; default plan_dirs to ["context/plans/backlog", "context/plans/active"]).
Always dry-run first:
```
python3 path/to/chevp-ai-framework/integration/github-issues/sync-plans-to-issues.py --dry-run --limit 3
```
Show the user 2–3 example issue titles + labels and confirm they look right before going wider.
Then run for the full set (without --limit) once the user approves the dry-run output.
For a single plan (e.g. "issue für §5.2"): use --limit 1 after temporarily reducing plan_dirs to the directory containing that plan, OR fall back to a direct gh issue create with the same title format the script would produce. Prefer the script for consistency.

Safety

This calls gh issue create and gh label create — both are user-visible writes to GitHub. Always confirm intent before the first non-dry-run invocation.
Never add --update without explicit user request — it overwrites existing issue bodies.
If the dry-run shows zero plans discovered, stop and investigate the plan_dirs config rather than assuming success.

Reference

Script + full docs: ../../integration/github-issues/README.md
Conceptual integration: ../../integration/kanban.md