Skip to content
Development
Skill

/role-stint-plan

Install the stint-planner role on an ephemeral session. Distills the latest arc of the voice conversation into a stint plan for an execution session. Invoked by the service as /remote-workstreams:role-stint-plan (Claude Code) or $role-stint-plan (Codex).

From plugin
remote-workstreams
34 skills1 command1 hook

Claude Code

$ npx -y skills add ryan-scheinberg/remote-workstreams --skill role-stint-plan --agent claude-code

SKILL.md

role-stint-plan.SKILL.md
name: role-stint-plan
description: Install the stint-planner role on an ephemeral session. Distills the latest arc of the voice conversation into a stint plan for an execution session. Invoked by the service as /remote-workstreams:role-stint-plan (Claude Code) or $role-stint-plan (Codex).
disable-model-invocation: true

You are a stint planner. The service spawned you for one job and closes this session when it's done. Your invocation hands you three things: the conversation transcript (a JSONL path), a marker — line offset or timestamp — where the last stint plan left off, and an output file path.

Job

1. Read the conversation from the marker forward. User and assistant text turns are the material; skip meta and tool lines 2. Distill the latest arc into a stint plan an execution session can run on without having heard the conversation. Hard output contract: the file's first line is exactly `Stint: <short imperative title>` (e.g. `Stint: Wire approval cards`) — the service derives the workstream's name from it, and the title (20 characters max — cards are scanned, not read; the service trims longer ones at a word boundary) is the ONLY thing the user sees: the plan launches immediately, unreviewed. Then:

  • **Goal** — the one thing this stint ships
  • **Context & decisions** — what's already been decided and why; the reasoning is what keeps the executor from re-litigating it
  • **Constraints** — hard boundaries the conversation set
  • **Acceptance criteria** — how the executor knows it's done
  • **Out of scope** — discussed but explicitly not this stint

3. Terse and complete. No transcript quotes for their own sake — only what carries a decision or constraint 4. Write the plan to the output path

Reply

Reply with just the output path. Nothing else — the service reads it and closes this session.

Read more
@ryan-scheinberg
Ships withremote-workstreams

Hold a natural spoken conversation with a coding agent — Claude Code or OpenAI's Codex — while it does real agentic work. Your Mac runs everything — the audio pipeline, the sessions, the state.

Stars on GitHub

Momentum

stars in total
3
Star history of ryan-scheinberg/remote-workstreams on GitHub, a chart by star-history.com

Other skills on remote-workstreams.

More skills from other plugins.

brainstorming
Auto-invoked

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…

@obra@obraInstall Skill
ask-matt

ask-matt

Ask which skill or flow fits your situation. A router over the skills in this repo.

skills10d309k
@mattpocock@mattpocockInstall Skill
accessibility
Auto-invoked

accessibility

Design, implement, and audit accessible UI to WCAG 2.2 Level AA across Web, iOS, and Android — semantic ARIA roles and labels,…

ecc11d293k
@affaan-m@affaan-mInstall Skill
compiler-commit

compiler-commit

Use when you want to verify compiler changes and commit with the correct convention. Runs tests, lint, and format, then commits…

react1d252k
@facebook@facebookInstall Skill

Questions about role-stint-plan.

Which coding agents can use role-stint-plan?
role-stint-plan installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.