plugins/v1tamins/skills/v1-bare-bones/SKILL.md
Use when a plan, PRD, proposal, or implementation outline is overscoped, too ambitious for the immediate goal, or needs to be reduced to a bare-bones version. Triggers on "bare bones", "no damn whistles", "no bells and whistles", "strip this plan", "trim this plan", "scope creep", "descope this plan", "MVP only".
npx skillsauth add v1-io/v1tamins v1-bare-bonesInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Strip an overscoped plan down to the smallest useful version.
Use this skill when the user wants less plan, less architecture, less ceremony, and no speculative extras. The job is not to make the plan more impressive. The job is to make it shippable.
Start from "Won't Have this time," then justify every promotion.
Promote an item to the bare-bones plan only when it passes at least one of these tests:
If an item is valuable but does not pass one of these tests, defer it. If it is ornamental, speculative, or included only because it seems nice, cut it.
Read the supplied plan, PRD, proposal, ticket, or implementation outline. If the plan references a specific local file, read that file. If it references nearby repo context and that context affects scope, inspect only the smallest relevant context.
Do not expand the plan while gathering context. Look for the original outcome, hidden dependencies, and evidence that a proposed item is truly required.
State the intended outcome in one sentence:
Outcome: [the smallest observable result this work needs to produce]
If the plan has multiple outcomes, choose the one the user asked for most directly. Move the others to deferred scope unless the user explicitly asked for a multi-outcome plan.
Classify each meaningful plan item:
| Bucket | Meaning | | --- | --- | | Keep | Required for first useful delivery | | Cut | Not needed, ornamental, duplicate, or speculative | | Defer | Valuable later, but not required now | | Question | Blocks trimming because the required outcome is ambiguous |
Keep the ledger concise. Group related items instead of listing every tiny subtask.
Challenge every Keep item:
If the answer is no, downgrade the item to Defer or Cut.
Produce a replacement plan with only these sections:
## Bare-Bones Plan
### Goal
[One sentence]
### Build
- [Smallest implementation step]
- [Smallest implementation step]
### Verify
- [Smallest meaningful validation]
### Non-Goals
- [Explicitly excluded item]
- [Explicitly excluded item]
### Blocking Questions
- [Only questions that must be answered before implementation]
Omit Blocking Questions when there are none. Do not include roadmap, future phases, rollout ceremony, dashboards, analytics, polish, or platform architecture unless one is required for the first useful delivery.
If the user asks to update a plan file, edit the existing artifact in place. Preserve the original heading style when possible, but replace bloated sections with the bare-bones plan.
If the user only asks for a review or trim, return the revised plan in chat and do not edit files.
Cut or defer these unless they pass the Must-have test:
Use this format:
Outcome: [one sentence]
## Bare-Bones Plan
### Goal
[one sentence]
### Build
- [...]
### Verify
- [...]
### Non-Goals
- [...]
### Blocking Questions
- [...]
## Scope Ledger
| Keep | Cut | Defer | Question |
| --- | --- | --- | --- |
| ... | ... | ... | ... |
Lead with the rewritten plan, not a long critique. Keep explanation short and concrete.
development
Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for large prs, new features/architectures, a deep code quality audit, or especially harsh maintainability review.
testing
Commit, push, open, and land a pull request through CI handoff. Use when work is complete and the user wants an agent to create or update a PR, open it as a draft, monitor GitHub checks with `gh pr checks`, fix failed checks, retry up to three remediation pushes, mark the PR ready for review once green, and move a linked Linear ticket to Human Review when one exists. Trigger on requests like 'land this PR', 'open and monitor a PR', 'commit push and watch CI', 'get this ready for review', or 'finish the PR workflow'.
development
Use when reviewing a PR, reviewing the current branch, or posting code review feedback to GitHub. Triggers on "review this PR", "code review", "check this pull request", "review my branch", "review and fix".
development
Use when panning recent merged PRs and agent session logs (Claude Code, Codex, Cursor) for compound-worthy moments to document, then queueing the gold through /ce-compound. Triggers on "goldpan", "pan for compound candidates", "pan for gems", "what should I /ce-compound", "find solutions to document", "scan sessions for learnings", "what did we learn last week", "compound candidates", "/v1-goldpan".