build-run-plan-flow

name

build-run-plan-flow

description

Configure the run-plan (implementation) phase of the /process workflow. Asks whether the user wants a run-plan step at all (skip-able), then asks whether they already have a run-plan process (e.g. an existing /run-plan command or skill that takes a plan doc and implements it). If they do, reads it and inspects whether it already commits/pushes/opens a PR; whatever is missing gets added. If they don't, bakes a simple inline implementation prompt directly into the wrapper. The ONLY file this skill writes is `.claude/skills/claude-slack-bridge_run-plan/SKILL.md` — a wrapper skill that runs the user's run-plan flow (`<@ref-run-plan-flow>`) or an inline implementation prompt when none exists, commits and pushes if the inner flow didn't, opens a GitHub PR, and sends a response back to the caller. Never scaffolds a separate `/run-plan` slash command. Returns a status of `configured` (with the captured reference) or `skipped` (with the literal label `run-plan-flow: skip`). Use as the run-plan-flow phase of /process-setup.

build-run-plan-flow — configure the /run-plan phase and write the claude-slack-bridge_run-plan wrapper

You run locally inside Claude Code (not via the Slack daemon). All clarifications go through AskUserQuestion. Do not call mcp__claude-slack-bridge__ask_on_slack from this skill — Slack is only the runtime channel for /process itself, not for setup.

This skill is the run-plan-flow phase of /process-setup. By the time it returns, either:

• The user opted in, the inputs were captured, and .claude/skills/claude-slack-bridge_run-plan/SKILL.md has been generated (status: configured); or
• The user opted out at Step 1 — no helper skill is written and the caller is told the literal label run-plan-flow: skip (status: skipped).

Return values the caller needs (printed as a fenced JSON block at the end of the final reply):

• status — configured or skipped.
• label — run-plan-flow: configured when configured, run-plan-flow: skip when skipped.
• has_existing_run_plan_process — true / false (only present when configured).
• existing_run_plan_reference — slash command or path to the user's existing run-plan flow (only when has_existing_run_plan_process == true).

Do not skip ahead to writing the helper skill until Step 3 has actually captured everything it needs.

---

Step 1 — does the user want a run-plan step at all?

Ask via AskUserQuestion:

Do you want a run-plan step in your /process workflow? (the run-plan step takes the approved plan doc, implements it in code, commits + pushes the changes, opens a PR for review, and only then hands off to the next step)

Options:

1. Yes, set up the run-plan step — continue to Step 2.
2. No, skip it — return immediately with status: "skipped" and label: "run-plan-flow: skip". Do not write any files.

---

Step 2 — does the user already have a run-plan process?

Capture the user's answer here so Step 3 knows which branch to take. Treat this as the working state of the skill — don't ask the same question twice in Step 3, just consume what you recorded here.

Ask via AskUserQuestion:

Do you already have a run-plan / implementation process for this repo? (an existing /run-plan slash command, a claude-slack-bridge_run-plan style skill, or any other repeatable flow that takes a plan doc and implements it in code)

Options:

1. Yes, I have one — record has_existing_run_plan_process = true. Then ask via AskUserQuestion (free-text reply via "Other"):

   Where is your run-plan process defined? Paste the slash command (e.g. /run-plan) or the relative path to the file (e.g. .claude/commands/run-plan.md, .claude/skills/run-plan/SKILL.md).

   Resolve the reply to a concrete file:

   • /foo → check .claude/commands/foo.md, then .claude/skills/foo/SKILL.md.
   • A path → use as-is.

   If neither file exists, ask once more whether the user meant a plugin command (in which case proceed with existing_run_plan_reference set to the slash command and a null file path) or wants to re-paste. Loop until resolved.

   Record existing_run_plan_reference (slash command form, e.g. /run-plan) and existing_run_plan_path (the resolved file path, or null for plugin commands). Continue to Step 3.

2. No, I don't have one — record has_existing_run_plan_process = false. The wrapper will bake in a simple inline implementation prompt that reads the plan doc and implements it. No further inputs needed here. Leave existing_run_plan_reference and existing_run_plan_path as null. Continue to Step 3.

---

Step 3 — write .claude/skills/claude-slack-bridge_run-plan/SKILL.md

This is the only file this skill ever writes. The wrapper's job at runtime is:

1. Implement the plan — either invoke the user's existing run-plan flow ({ref_run_plan_flow}), or, when there is no existing flow, run a simple inline implementation prompt baked directly into this same skill.
2. Commit and push if the inner flow didn't already — git add -A, git commit -m "run-plan: <feature>", git push -u origin <branch>. Skip this if the inner flow already committed + pushed.
3. Open a GitHub PR — gh pr create --base main --head <branch> --title "<feature>: implementation" --body "...". If a PR already exists for the branch, update it instead.
4. Send a response back to the caller — post via mcp__claude-slack-bridge__ask_on_slack with the PR URL and a short summary so the workflow engine can route to the next step.

Inspect the existing flow (only when has_existing_run_plan_process == true) to decide which extra steps the wrapper has to add:

• Does the inner flow already git commit + git push? If not, the wrapper must do step 2.
• Does the inner flow already gh pr create / open a PR? If not, the wrapper must do step 3.
• Does the inner flow already respond via ask_on_slack? If not, the wrapper must do step 4.

Whatever the inner flow already does, the wrapper should NOT duplicate — it just fills the gaps. Whatever it doesn't do, the wrapper must add. If has_existing_run_plan_process == false, treat all three boolean facts as false — the wrapper does steps 2–4 itself, and step 1 runs an inline implementation prompt.

Write the wrapper

Create .claude/skills/claude-slack-bridge_run-plan/ if missing, then write SKILL.md (atomic write: .SKILL.md.tmp → os.replace) using the template below. Substitute the placeholders inline (do not leave them literal in the output):

• {step_1_body} → see the two variants below, picked based on has_existing_run_plan_process.
• {inner_does_commit_push}, {inner_does_pr}, {inner_does_respond} → boolean facts captured during inspection (all false when there is no existing flow). Use them to phrase the wrapper steps as "skip if already done by the inner flow" vs. "always do".

Variant A — has_existing_run_plan_process == true (substitute {ref_run_plan_flow} with the slash command captured in Step 2, e.g. /run-plan):

Invoke `{ref_run_plan_flow}` to implement the approved plan. The plan doc lives under `.plan/<feature>.md` (path from `.claude/process.json`). Pass the plan path to the inner flow and let it apply the code changes.

Variant B — has_existing_run_plan_process == false (simple inline implementation prompt, no extra config from the user):

Implement the approved plan inline (there is no separate `/run-plan` command — this wrapper is the whole flow):

1. Read `.claude/process.json` to get the feature slug, branch, and plan path (defaults to `.plan/<feature>.md`).
2. Read the plan doc end-to-end so you have the full task list and any constraints in mind.
3. Apply the code changes the plan describes, one section at a time. Run tests / type checks / linters as they would normally run in this repo after meaningful edits.
4. Stop when every item in the plan is implemented. The next steps commit, push, open a PR, and respond.

Now write the wrapper file using this template:

---
name: claude-slack-bridge_run-plan
description: "Implement the approved plan for the current feature, commit + push the code changes, open a GitHub PR for review, and post the PR URL back to the caller via mcp__claude-slack-bridge__ask_on_slack. Either wraps an existing run-plan flow (filling in whatever steps it doesn't already perform) or runs an inline implementation prompt baked into this skill."
---

# claude-slack-bridge_run-plan — run-plan phase wrapper

This skill is invoked by the `/process` workflow engine as the run-plan step. It is the single entry point the engine calls; everything the run-plan phase needs to do at runtime is encoded here.

## 1. Implement the plan

{step_1_body}

## 2. Commit and push

If the inner flow already committed and pushed the changes, skip this step. Otherwise:

git add -A git commit -m "run-plan: " git push -u origin

Use the branch from `.claude/process.json` (`feature/<slug>` by default).

## 3. Open a GitHub PR

If the inner flow already opened a PR, skip and capture the PR URL it returned. Otherwise:

- Run `gh pr view <branch>` to check for an existing PR.
- If none: `gh pr create --base main --head <branch> --title "<feature>: implementation" --body "<short summary of what was implemented, derived from the plan doc>"`.
- If one exists: `gh pr edit <PR#> --body "<refreshed summary>"` and let the push from step 2 surface the new commits.

Capture the resulting PR URL.

## 4. Respond to the caller

Post the result via `mcp__claude-slack-bridge__ask_on_slack` so the workflow engine can route to the next step:

> Implementation ready for review: <PR-URL>. Reply `approve` to continue to the next step, or leave review comments on the PR and reply `comments` here when done.

- `approve` → return `status = approved`; the engine advances to the next step.
- `comments` → return `status = needs_fixes` with the PR URL; the engine routes to `/required-fixes` before resuming.

Do not return until the user has replied.

After writing the file, briefly confirm to the user via AskUserQuestion:

I wrote .claude/skills/claude-slack-bridge_run-plan/SKILL.md ({wraps {ref_run_plan_flow} | runs an inline implementation prompt}, fills in the missing commit/push/PR/respond steps). Look right?

Options:

1. Yes — return with status: "configured".
2. No, fix it — ask what to change (wrong existing_run_plan_reference? inner-flow inspection got it wrong? switch to the inline prompt?), update the captured values, re-write the file, ask again. Cap retries at ~3.
3. Skip — delete the half-written file and return with status: "skipped" + label: "run-plan-flow: skip".

---

Return shape

End your final reply with a single fenced JSON block so the caller can parse the result.

For the configured case (user had an existing process, or the wrapper now contains an inline implementation prompt):

{
  "status": "configured",
  "label": "run-plan-flow: configured",
  "has_existing_run_plan_process": true,
  "existing_run_plan_reference": "/run-plan"
}

When has_existing_run_plan_process == false, existing_run_plan_reference is null:

{
  "status": "configured",
  "label": "run-plan-flow: configured",
  "has_existing_run_plan_process": false,
  "existing_run_plan_reference": null
}

For the skipped case (Step 1 returned No):

{
  "status": "skipped",
  "label": "run-plan-flow: skip"
}