shape
Skill2 starsChanged 9 days agoMITView on GitHub
Turn the owner's plan into dragonhoard milestones and self-contained task packs (divide and conquer). Proposes the milestone cut first and writes nothing until the owner confirms. Also writes the packs for the next milestone when it becomes current. Use when the user says "shape this plan", "break this plan down", "make milestones", or invokes /dragonhoard:shape [plan file | milestone ID].
36 other projects use a version of this.
Adopted by
Most recently taken up by
- jskswamy/skillstheir own version, first seen Nov 8, 2026
- Belweave/hstacktheir own version, first seen Nov 8, 2026
- Mrbaeksang/skillstheir own version, first seen Nov 8, 2026
- chrislacey89/skillstheir own version, first seen Nov 4, 2026
- theexperiencecompany/gaiatheir own version, first seen Nov 3, 2026
- jayden-dang/skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use layout, build and 9 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
owshq-mecSkilllayout
Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.
Also in 8 projects that use this
Microsoft AzureSkillbuild
Build the TypeSpec Rust emitter from source. Run this after cloning the repo, after pulling changes, or when TypeScript source files have been modified.
Also in 5 projects that use this
AnthropicSkillbrief
Generate contextual briefings for legal work — daily summary, topic research, or incident response. Use when starting your day and need a scan of legal-relevant items across email, calendar, and contracts, when researching a specific legal question across internal sources, or when a developing situation (data breach, litigation threat, regulatory inquiry) needs rapid context.
Also in 3 projects that use this
jaydubya818Skillwhy
Use when design rationale must be recovered from source history and available records.
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 7 projects that use this
SentrySkillspec
Produce a verified spec — problem, desired outcome, and acceptance criteria — before building. Use when given a GitHub/Linear issue link, when starting from a fuzzy in-conversation idea, or asked to "figure out what to do" / "scope this" / "what should I build" — whenever the *what* isn't yet pinned down. Grills the ambiguities, checks the premise against the code, and produces acceptance criteria you sign off on — then hands to design-first.
Also in 3 projects that use this
Other versions36 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 36
- hoophq/hoop★ 848same name
- educlopez/ui-craft★ 441same name
- 10xChengTu/Mux0★ 337same name
- theexperiencecompany/gaia★ 325same name
- jagrat7/linux-wallpaper-engine★ 227same name
- owshq-mec/semana-ai-data-engineer★ 122same name
- StyleSwift/StyleSwift★ 61same name
- tdimino/claude-code-minoan★ 41same name
- saltandsilicon/game-theorist★ 16same name
- excatt/superclaude-plusplus★ 14same name
26 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 3 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 34 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 42
Ask the owner to approve or adjust, and wait for the answer (in Claude Code, use AskUserQuestion).
- Distinctive: line 61
Every pack must be self-contained:
- Very specific: line 71
Write
STATUS.md: Add one row per task across all milestones, in execution order: | # | M<n>-T<nn> | title | TODO | deps | pack |. - Very specific: line 88
Run this when dragonhoard.py next reaches a task with no pack.
- Very specific: line 90
Read <root>/M<n>-<slug>/PLAN.md, DIGEST.md, LESSONS.md, the previous milestone's GATE.md results, and the option.chosen events that affect …
- Distinctive: line 93
If the tasks no longer fit (scope, order or contracts changed), propose the change and ask the owner to confirm it before writing anything.
- Very specific: line 95
Write the packs as in A.4.
“Common” is measured: kept by at least 6 in 10 of the 34 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
dragonhoard shape
CLI: python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dragonhoard.py" (called dragonhoard.py below).
Arguments: $ARGUMENTS. That is a plan file path, a milestone ID such as M2, or empty (then ask
for the plan).
Requires dragonhoard.json. If it is missing, run /dragonhoard:init first. <root> is its root
value.
A. First shape (a plan file or pasted plan)
- Read the whole plan. Also read
<root>/CONSTITUTION.md,<root>/README.md,<root>/LESSONS.mdand the skeletons in<root>/templates/. If the project already has code, run one scout to map the existing structure (dragonhoard.py role scoutsays how). Do not read it all yourself. Propose the milestone cut, and write nothing yet. Present
M0..Mn. For each milestone give:- its goal, as one observable sentence;
- why it comes in this order (what it unblocks);
- its gate criterion (how we know it works end to end);
- its rough tasks (titles only).
Rules for the cut:
- M0 is foundation: the project runs, the checks run, and the first test passes.
- Each milestone ends in something runnable.
- Contracts and interfaces come before their consumers.
- Risky unknowns come early, as tasks with Open options.
- Anything the plan calls "later" or "nice to have" goes into a final milestone, or out of scope.
Ask the owner to approve or adjust, and wait for the answer (in Claude Code, use AskUserQuestion). Repeat until approved. Log the approved cut:
dragonhoard.py log - shape.approved "<n> milestones: M0 <name>, M1 <name>, ..."- Write the milestone folders. For every milestone, create
<root>/M<n>-<slug>/. Copy the files in<root>/templates/phase/into it and fill them in:PLAN.md: goal, why, a task table, owner decisions, out of scope.CONSTITUTION.md: only the rules specific to this milestone, such as its contracts, boundaries and traps. Leave the article list empty rather than invent rules.PROGRESS.md: oneTODOrow per task, with the planned files in backticks.GATE.md: the gate commands and the criteria checklist.
Write task packs for the first milestone only. Copy
<root>/templates/TASK.mdto<root>/M0-<slug>/tasks/M0-T<nn>-<slug>.mdfor each task. Later milestones get packs when they become current (section B). Packs written early go stale.Every pack must be self-contained:
- Context to load names the exact files, and copies the plan excerpts the task needs, with section references. A coding agent must never need to open the whole plan.
- Files: 1–3 implementation files plus one focused test. Split anything bigger.
- Consumes points at real
path:linelocations for code that exists. For code an earlier task will produce, give the task ID and the planned signature. - Open options lists every unresolved choice: library, storage, algorithm, API shape. Those choices get explored in the loop instead of guessed.
- Acceptance lists observable checks, never "works well".
- Write
STATUS.md. Add one row per task across all milestones, in execution order:| # | M<n>-T<nn> | title | TODO | deps | [pack](M<n>-<slug>/tasks/...) |. Use—for the pack while it is unwritten. Add one row per gate:| G<n> | not started | [GATE](M<n>-<slug>/GATE.md) |. The first task of milestone M<n>, for n ≥ 1, depends onG<n-1>. Verify and hand off.
dragonhoard.py check # must PASS: STATUS and PROGRESS agree dragonhoard.py next # must name M0-T01 dragonhoard.py digestReport the milestone list and the first task. Then suggest
/dragonhoard:next.
B. Next milestone (/dragonhoard:shape M<n>)
Run this when dragonhoard.py next reaches a task with no pack.
- Read
<root>/M<n>-<slug>/PLAN.md,DIGEST.md,LESSONS.md, the previous milestone'sGATE.mdresults, and theoption.chosenevents that affect this milestone. Reality has moved since the first shape; the packs must reflect what was actually built. - If the tasks no longer fit (scope, order or contracts changed), propose the change and ask the owner to confirm it before writing anything.
- Write the packs as in A.4. Fill the
[pack]links inSTATUS.mdand the milestone'sPLAN.md. Logdragonhoard.py log - shape.milestone "M<n>: <k> packs", then runcheck.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

