Agent skill
pick-next-prompt
Pick the next ready shell from a prompt plan in .turbo/prompt-plans/ and hand it to /turboplan for fill-in. Use when the user asks to "pick next prompt", "next prompt", "continue prompt plan", "what's next", "next implementation step", or "continue with the plan".
Install this agent skill to your Project
npx add-skill https://github.com/tobihagemann/turbo/tree/main/skills/pick-next-prompt
SKILL.md
Pick Next Prompt
Pick the next ready shell from a prompt plan index at .turbo/prompt-plans/<slug>.md and hand it to /turboplan in shell mode.
Step 1: Resolve the Prompt Plan Index
Determine which prompt plan index to read using these rules in order:
- Explicit path — If the user passed a file path, use it
- Explicit slug — If a slug was passed, resolve to
.turbo/prompt-plans/<slug>.md - In-progress wins — Glob
.turbo/prompt-plans/*.md. If exactly one index has any prompt withStatus: in-progress, use it. This is the prompt plan currently being worked on. - Single file — If exactly one index file exists, use it
- Most recent — If no index has an in-progress prompt and rule 3 did not resolve, use the most recently modified file
- Legacy fallback — If
.turbo/prompt-plans/does not exist but.turbo/prompts.mdexists, tell the user to re-run/draft-prompt-planto upgrade to shells, and halt. - Nothing found — If no prompt plan exists, tell the user to run
/turboplanfor a complex task (which will route to/draft-spec+/draft-prompt-plan) and stop
If multiple files have in-progress prompts (concurrent work in different feature branches), use AskUserQuestion to let the user pick.
State the resolved index path before continuing.
Read the index file. Each prompt entry uses bold-prefixed inline markers under a ## Prompt N: heading. Parse:
- Prompt number and title (from the
## Prompt N: <title>heading) - Status (
**Status:** pending/in-progress/done) - Shell file path (
**Shell:** <path>) - Dependencies (
**Depends on:** noneorPrompt N)
Step 2: Pick the Next Ready Shell
Find the first pending prompt whose dependencies are all done.
- If found — proceed to Step 3
- If all prompts are
done— report completion to the user — the plan is finished - If remaining prompts are blocked — report which prompts are blocked and by what
Step 3: Mark In-Progress and Read the Shell
Update the index file to mark the selected prompt in-progress.
Read the shell file at the path from the prompt's Shell: field. Verify the file exists and contains the expected shell structure (Context, Produces, Consumes, Covers Spec Requirements, Implementation Steps, Open Questions). If the shell file is missing, stop and report — the index and shell files are out of sync.
Step 4: Run /turboplan Skill
Run the /turboplan skill, passing the shell file path as input.
Tell /turboplan (via its task description) that the plan's final implementation step must include: "Mark prompt N as done in the prompt plan index at <index path>."
Rules
- Never modify the spec file.
- Do not pre-verify Consumes or refresh surveys here.
- Do not re-draft or re-analyze the shell.
- If the index references a shell file that does not exist, halt. Do not silently recover by regenerating the shell.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
review-api-usage
Check API, library, and framework usage in code against official documentation and installed skill knowledge. Flags deprecated APIs, incorrect method signatures, wrong parameter types, version-incompatible patterns, and best-practice violations. Use when the user asks to "review API usage", "check API usage", "verify against docs", "check library usage", "validate API calls", "check against documentation", or "check for deprecated APIs".
resolve-pr-comments
Evaluate, fix, answer, and reply to GitHub pull request review comments. Handles both change requests (fix or skip) and reviewer questions (explain using reasoning recalled from past Claude Code transcripts). Use when the user asks to "resolve PR comments", "fix review comments", "address PR feedback", "handle review comments", "address review feedback", "respond to PR comments", "answer review questions", or "address code review".
consult-codex
Multi-turn consultation with Codex CLI for second opinions, brainstorming, or collaborative problem-solving. Use when the user asks to "consult codex", "ask codex", "get codex's opinion", "brainstorm with codex", "discuss with codex", or "chat with codex".
review-tooling
Detect what dev tooling infrastructure a project has and flag gaps across linters, formatters, pre-commit hooks, test runners, and CI/CD pipelines. Returns structured findings without applying changes. Use when the user asks to "review tooling", "check project tooling", "what tooling is missing", "review dev infrastructure", or "tooling audit".
create-changelog
Create a CHANGELOG.md following keepachangelog.com conventions with version history backfilled from GitHub releases or git tags. Use when the user asks to "create a changelog", "add a changelog", "initialize changelog", "start a changelog", "set up changelog", "generate changelog", or "backfill changelog".
update-changelog
Update the Unreleased section of CHANGELOG.md based on current changes. No-op if CHANGELOG.md does not exist. Use when the user asks to "update changelog", "add to changelog", "update the changelog", "changelog entry", "add changelog entry", or "log this change".
Didn't find tool you were looking for?