create-work-item
Scaffold a new developer-scoped project directory under work/ that follows the work-directory conventions (README + launcher next-session.md + ledger…
Skill by kashman001
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/kashman001-create-work-item
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/create-work-item
What's inside
- Skill name
create-work-item- Description
- Scaffold a new developer-scoped project directory under work/ that follows the work-directory conventions (README + launcher next-session.md + ledger handoff.md). Use when starting a new piece of multi-session work, when the user says "create a work item / work directory / project", or before beginning work that will span more than one session and needs a launcher + ledger. For single-file or one-shot tasks, do not create a work directory.
- License
- MIT
- Instructions
- 170 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
## Purpose
Create a `work/<project-name>/` directory pre-wired with the standard backbone
files (README, launcher, ledger) so multi-session work starts consistent and
stays context-cheap. This skill owns the scaffolding procedure; the durable
definition of each file's role lives in `docs/work-directory-conventions.md`.
Do **not** use this for trivial single-step tasks — those need no work
directory. Use it when work will span multiple sessions or needs persistent,
agent-neutral state on disk.
## Required Reads
**Always read first:**
1. `docs/work-directory-conventions.md` — the file roles and write discipline.
**Read on demand:**
1. `skills/session-rollover/SKILL.md` — how the launcher and ledger are
maintained (and rolled over) across sessions.
2. `skills/decision-log/SKILL.md` — only if the project will keep a
`decisions.md` / ADRs.
## Procedure
1. **Confirm scope and name.** Pick a hyphenated directory name
(`work/<project-name>/`). Ask the user for the governing skill(s) if any
workflow skill applies; use `—` if none.
**Ask whether the effort needs a spec.** Rule of thumb: a spec is required
when success criteria can't fit the README's one-line goal — multi-feature
work, user-facing behaviour, anything where "done" is genuinely debatable.
Ops-flavoured efforts skip it. An external tracker ticket (JIRA, GitHub
issue) with real acceptance criteria can *be* the spec of record. Skeleton
and full rules: `docs/agents/issue-tracker.md` → "Spec conventions".
**If no spec: the README's `## Success criteria` section is required.**
2. **Create the directory and the three backbone files** below. Substitute
`<project>`, a one-line description, the governing skill(s), and the first
objective. Use the canonical names `next-session.md` (launcher) and
`handoff.md` (ledger).
3. **README.md** (project identity — durable, changes rarely):
```markdown
# <Project> — <one-line what-it-is>
Governing skill(s): `<skill>` (or none).
**Start here:** `next-session.md` (catch-up launcher) → `handoff.md`
(session ledger, top block).
## What this is
<2–4 sentences: goal, scope, why it exists.>
## Success criteria
<!-- Required when the effort has no spec.md; delete if spec.md exists. -->
<How we'll know this effort is done — verifiable, not "make it work".>
## Files
- `next-session.md` — forward launcher (what to do next). REPLACED each rollover.
- `handoff.md` — session ledger (what happened). APPEND newest-on-top; archive
to `handoff-archive.md` when it exceeds the two most recent sessions.
- <other project files, added as they appear.>
```
4. **next-session.md** (launcher — starts nearly empty, forward-only):
```markdown
# Catchup prompt — <Project> (paste into a new agent session)
We're resuming <project>. Works in any runtime (Claude Code, Codex, Gemini,
OpenCode) — all read `CONTEXT.md` via their entrypoint.
> **This file is the LAUNCHER (catch-up prompt).** Forward-only, and REPLACED
> at each rollover: it holds what to do next, still-binding constraints, and
> pointers — never session history. Past-tense provenance lives in
> `handoff.md` (the append-only ledger). Convention:
> docs/work-directory-conventions.md.
## >>> START HERE <<<
<The current objective and the concrete next actions (a short numbered todo).>
## Constraints already decided (do not re-litigate)
<One-liners + pointers, added as decisions are made.>
## Read these first, in order
1. `work/<project>/README.md`
2. `work/<project>/handoff.md` (top block)
```
**Plan markers — only when asked** (`--plan` among the arguments, or the
user says the effort will run a plan per `docs/plans.md`); the default
scaffold carries none, plans stay opt-in. Append to `next-session.md`:
```markdown
## Position
<!-- plan:begin position -->
<!-- plan:end position -->
```
`scripts/plan.sh sync` renders the plan's position and frontier between
the markers and refuses a launcher without them.
5. **handoff.md** (ledger — purpose header + first block):
```markdown
<!--
PURPOSE: This is the LEDGER (provenance log). Append-only, newest block on
TOP. Each "# Session Handoff" block records what happened in one session.From kashman001/ai-workspace-template, MIT.
More from kashman001/ai-workspace-template
- 0.570score
constraint-driven-development
kashman001
Write a project's quality bar into CONSTRAINTS.md and flag changes that quietly lower it (new suppressions, skipped tests, stripped assertions, thresholds…
SkillAFreshness: AClaude Code +9 - 0.565score
doubt-driven-development
kashman001
Subject each non-trivial decision to a fresh-context adversarial review before it stands.
SkillAFreshness: AClaude Code +9 - 0.561score
doc-review
kashman001
Multi-perspective review of a technical document: audience gate first, six independent reviewer subagents, findings synthesized into prioritized…
SkillAFreshness: AClaude Code +9 - 0.560score
decision-log
kashman001
Capture the why behind a decision (choice, rejected alternative, constraint) as a commit trailer, a Tier-2 note in work/<project>/decisions.md, or a promoted…
SkillAFreshness: AClaude Code +9 - 0.557score
checkpoint
kashman001
Wrap up a finished chunk of work: reconcile the backlog, memory and docs, write the hand-off, confirm a clean branch, and emit a catch-up prompt for the next…
SkillAFreshness: AClaude Code +9 - 0.557score
design-for-testability
kashman001
Advisory six-question interrogation of a design: how will we test it, how does it fail, what is observable, and what is the cheapest check that proves it works.
SkillAFreshness: AClaude Code +9 - 0.544score
git-guardrails-claude-code
kashman001
Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute.
SkillAFreshness: AClaude Code +9 - 0.544score
diagnosing-bugs
kashman001
Diagnosis loop for hard bugs and performance regressions.
SkillAFreshness: AClaude Code +9 - 0.541score
create-verification-skill
kashman001
Generate a project-local verification skill that drives your app the way a user does — any language, framework, or platform.
SkillAFreshness: AClaude Code +9 - 0.539score
benchmark-checklist
kashman001
Vet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it.
SkillAFreshness: AClaude Code +9 - 0.539score
codebase-design
kashman001
Shared vocabulary for designing deep modules.
SkillAFreshness: AClaude Code +9 - 0.539score
domain-modeling
kashman001
Build and sharpen a project's domain model.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 12, 2026 (17 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from kashman001
- 0.570score
constraint-driven-development
kashman001
Write a project's quality bar into CONSTRAINTS.md and flag changes that quietly lower it (new suppressions, skipped tests, stripped assertions, thresholds…
SkillAFreshness: AClaude Code +9 - 0.565score
doubt-driven-development
kashman001
Subject each non-trivial decision to a fresh-context adversarial review before it stands.
SkillAFreshness: AClaude Code +9 - 0.561score
doc-review
kashman001
Multi-perspective review of a technical document: audience gate first, six independent reviewer subagents, findings synthesized into prioritized…
SkillAFreshness: AClaude Code +9 - 0.560score
decision-log
kashman001
Capture the why behind a decision (choice, rejected alternative, constraint) as a commit trailer, a Tier-2 note in work/<project>/decisions.md, or a promoted…
SkillAFreshness: AClaude Code +9 - 0.557score
checkpoint
kashman001
Wrap up a finished chunk of work: reconcile the backlog, memory and docs, write the hand-off, confirm a clean branch, and emit a catch-up prompt for the next…
SkillAFreshness: AClaude Code +9 - 0.557score
design-for-testability
kashman001
Advisory six-question interrogation of a design: how will we test it, how does it fail, what is observable, and what is the cheapest check that proves it works.
SkillAFreshness: AClaude Code +9
Cite this page
create-work-item skill. Agent Mart. https://agentmart.dev/skills/kashman001-create-work-item (last updated 2026-10-24)