How do I install the Seeds Issue Workflow skill?

Run npx skillmds@latest add jayminwest/seeds-issue-workflow in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Seeds Issue Workflow safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Seeds Issue Workflow?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Seeds Issue Workflow free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Seeds Issue Workflow?

jayminwest (@jayminwest) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Seeds Issue Workflow by jayminwest — Agent Skill

Name: Seeds Issue Workflow
Author: jayminwest

Seeds Issue Workflow

Turn a request into tracked, ready seeds work — file an issue with priority and labels, wire dependencies, and decompose large work into structured child seeds with sd plan. Use when an agent needs to create or organize work in the seeds repo.

jayminwest d4f1f77 2.8 KB Updated Sep 17, 2026

File contents

seeds-issue-workflow

Use this skill inside the seeds repo when you need to turn a vague ask into a tracked unit of work that sd ready will surface. It encodes two procedures: create-then-ready for a single scoped task, and plan-decomposition for work large or ambiguous enough that an LLM benefits from structured breakdown first.

Pre-flight

sd prime                          # inject rules + command reference
sd ready                          # see what is already unblocked
sd search "<keywords>"            # avoid filing a duplicate

If a seed already covers the work, claim it (sd update <id> --status in_progress) instead of filing a new one.

Procedure A — create-then-ready (single scoped task)

For a task that fits in one focused commit:

sd create --title "Fix sd close race on concurrent writers" \
  --type bug --priority 1 \
  --description "Two agents closing the same id can drop one update."

Capture the printed id (e.g. seeds-1a2b), then tag and wire it:

sd label add seeds-1a2b concurrency
sd dep add seeds-1a2b seeds-9f8e   # this work depends on seeds-9f8e first

A seed is ready when it is open and every id in its blockedBy list is closed. Add dependencies for real ordering only — every unnecessary sd dep add hides the work from sd ready. Verify:

sd ready | grep seeds-1a2b        # appears once its blockers close

Procedure B — plan-decomposition (large / ambiguous work)

When the work is multi-step, emit a structured prompt, fill it, and let seeds spawn the children:

sd plan prompt seeds-1a2b > /tmp/plan.json   # emit the template to fill
$EDITOR /tmp/plan.json                       # write one step per unit of work
sd plan submit seeds-1a2b --plan /tmp/plan.json

Each step becomes one child seed; a step's blocks: [j] wires step i's id into step j's blockedBy, so the children come out in dependency order. Steps may carry labels: [...] that flow to the spawned child (normalized and merged additively). Inspect and track:

sd plan show <pl-id>              # sections, children, sub-plans
sd ready                          # the first unblocked child surfaces

Finish

Close completed work and sync before handing off:

sd close seeds-1a2b --reason "Fixed; covered by store.test.ts"
sd sync                           # stage + commit .seeds/ changes

Do not git push unless the user asks. Never hand-edit .seeds/issues.jsonl — always go through sd so advisory locks and atomic writes are honored.

---name: seeds-issue-workflowdescription: Turn a request into tracked, ready seeds work — file an issue with priority and labels, wire dependencies, and decompose large work into structured child seeds with sd plan. Use when an agent needs to create or organize work in the seeds repo.---# seeds-issue-workflowUse this skill inside the seeds repo when you need to turn a vague askinto a tracked unit of work that `sd ready` will surface. It encodestwo procedures: **create-then-ready** for a single scoped task, and**plan-decomposition** for work large or ambiguous enough that an LLMbenefits from structured breakdown first.## Pre-flight```bashsd prime # inject rules + command referencesd ready # see what is already unblockedsd search "<keywords>" # avoid filing a duplicate```If a seed already covers the work, claim it(`sd update <id> --status in_progress`) instead of filing a new one.## Procedure A — create-then-ready (single scoped task)For a task that fits in one focused commit:```bashsd create --title "Fix sd close race on concurrent writers" \ --type bug --priority 1 \ --description "Two agents closing the same id can drop one update."```Capture the printed id (e.g. `seeds-1a2b`), then tag and wire it:```bashsd label add seeds-1a2b concurrencysd dep add seeds-1a2b seeds-9f8e # this work depends on seeds-9f8e first```A seed is **ready** when it is open and every id in its `blockedBy`list is closed. Add dependencies for real ordering only — everyunnecessary `sd dep add` hides the work from `sd ready`. Verify:```bashsd ready | grep seeds-1a2b # appears once its blockers close```## Procedure B — plan-decomposition (large / ambiguous work)When the work is multi-step, emit a structured prompt, fill it, and letseeds spawn the children:```bashsd plan prompt seeds-1a2b > /tmp/plan.json # emit the template to fill$EDITOR /tmp/plan.json # write one step per unit of worksd plan submit seeds-1a2b --plan /tmp/plan.json```Each step becomes one child seed; a step's `blocks: [j]` wires step`i`'s id into step `j`'s `blockedBy`, so the children come out independency order. Steps may carry `labels: [...]` that flow to thespawned child (normalized and merged additively). Inspect and track:```bashsd plan show <pl-id> # sections, children, sub-planssd ready # the first unblocked child surfaces```## FinishClose completed work and sync before handing off:```bashsd close seeds-1a2b --reason "Fixed; covered by store.test.ts"sd sync # stage + commit .seeds/ changes```Do not `git push` unless the user asks. Never hand-edit`.seeds/issues.jsonl` — always go through `sd` so advisory locks andatomic writes are honored.

Frequently asked questions