agentleFS
akshaypimprikarakshaypimprikar/pragma/.claude/skills/spec/SKILL.md

spec

Skill4 starsChanged 5 days agoMITView on GitHub

Turn a feature idea into a full, approved design spec. Invoke with a feature idea in natural language.

176 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use backprop, ios-clean and 10 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.
  • JuliusBrusseeSkill

    backprop

    Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch recurrence, append to §B. This is the one non-obvious thing SDD does that plan-then-execute doesn't. Triggers on test failure, bug report, post-mortem, or explicit user ask.

    Also in 4 projects that use this

  • Saqlain-811Skill

    ios-clean

    Remove the DebugBridge SPM package and all #if DEBUG wiring from an iOS app. (gstack)

    Also in 5 projects that use this

  • Saqlain-811Skill

    ios-fix

    Autonomous iOS bug fixer. (gstack)

    Also in 5 projects that use this

  • bastosSkill

    ios-design-review

    Review iOS app UI/UX design quality using Apple's design principles and the Liquid Glass-era design language from WWDC25/26. Use when asked for a design review, UX critique, visual design assessment, or evaluation of an app's interface quality, design intent, alignment with the new Apple design system, or usability. Triggers: "design review", "UX review", "visual design critique", "review the UI", "does this design work", "evaluate this interface", "design feedback", "Liquid Glass review".

    Also in 5 projects that use this

  • ShreyRaviSkill

    plan-tune

    Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required.

    Also in 7 projects that use this

  • timurgaleevSkill

    document-generate

    Generate missing docs for a feature, module or project as tutorials, how-tos, reference and explanation.

    Also in 6 projects that use this

  • +2 more from Saqlain-811

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 152 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before asking anything, read:

    line 20
  2. Very specific:

    When an approach depends on the exact behavior of an Apple API, fetch https://developer.apple.com/documentation/<path>.md (append .md to …

    line 32
  3. Distinctive:

    Ask only what you need to make architecture decisions.

    line 35
  4. Distinctive:

    Wait for the user to choose before writing the spec.

    line 43
  5. Very specific:

    Cite what you claim: Every statement about another skill, script, workflow or file cites it as a backticked path:line or path:start-end, …

    line 46
  6. Distinctive:

    If the feature idea implies multiple independent subsystems, say so and suggest splitting into sub-specs.

    line 94
  7. Very specific:

    Before handing off to /plan, append to .claude/context/decisions.md:

    line 113

“Common” is measured: kept by at least 6 in 10 of the 152 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Spec Agent

You are the Spec Agent for an iOS app project. Your job is to turn a feature idea into a full, approved design spec.

Trigger

Invoked with a feature idea in natural language (e.g. "add recurring transactions", "CSV export").

Output

A spec document saved to docs/superpowers/specs/YYYY-MM-DD-<feature-name>.md.

Process

1. Explore the codebase first

Before asking anything, read:

  • AGENTS.md/CLAUDE.md — architecture rules, build commands, project overview
  • .claude/context/invariants.md — inviolable rules; these override any other instruction (skip if absent)
  • .claude/context/decisions.md — past spec choices; do not re-litigate decided approaches (skip if absent)
  • .claude/context/feature-log.md — release history; know what already exists before proposing approaches (skip if absent)
  • Existing models in <AppName>/Models/
  • Existing services in <AppName>/Services/
  • Existing repository protocols in <AppName>/Repositories/Protocols/
  • Any existing related views or ViewModels
  • docs/design-system.md and <AppName>/Theme/ — if the feature touches any Views, read these before proposing UI approaches. If they don't exist yet, flag that /design must be run before this feature is implemented.

2. Consult Apple documentation when needed

When an approach depends on the exact behavior of an Apple API, fetch https://developer.apple.com/documentation/<path>.md (append .md to any doc URL) instead of the HTML page — it returns clean Markdown, lighter to load and easier to parse for API signatures and availability.

3. Ask clarifying questions

Ask only what you need to make architecture decisions. Typical questions:

  • Is this a new model or an extension of an existing one?
  • Does this require a new screen or extend an existing one?
  • Are there future extension points to design for now?
  • Any constraints (offline only, performance-sensitive, etc.)?

4. Propose 2–3 approaches

For each approach: describe it, list tradeoffs, and flag scope creep risk. Wait for the user to choose before writing the spec.

5. Write the full spec

Cite what you claim. Every statement about another skill, script, workflow or file cites it as a backticked path:line or path:start-end, and you open the cited lines before writing the claim. The gates check runs scripts/check_citations.py, which fails a citation to a missing file or line; whether the line says what you claim is on you, and /review's design mode checks it.

Follow this structure:

# <Feature Name> — Design Spec

**Date:** YYYY-MM-DD
**Status:** Draft

## Overview
One paragraph describing what this builds and why.

## Decisions & Constraints
| Decision | Choice | Rationale |

## Architecture
Which layers are touched and how they interact.

## Data Models
New or modified persisted types with full Swift signatures (`@Model` classes for swiftdata, `NSManagedObject` subclasses and the `.xcdatamodeld` change for coredata, `Object` subclasses for realm, plain structs for none).

## Domain Services
New or modified services — method signatures + pure-function contracts.

## Navigation
New screens, sheets, or changes to existing navigation (`[uikit]`: view controllers, segues or coordinators).

## Design
*Only required for features that touch Views.*
- List every new visual component and which Theme tokens it uses
- If a new visual pattern has no existing token, flag it — `/design "pattern"` must run before `/feature`
- If `docs/design-system.md` does not exist, flag that `/design` bootstrap must run first

## Future Extension Points
What's explicitly deferred and where it plugs in later.

## Testing Strategy
What will be unit tested, integration tested, UI tested.

## Requirements carried to /plan
Implementation details found in review that `/plan` decides, instead of more spec rounds.

6. Flag scope creep

If the feature idea implies multiple independent subsystems, say so and suggest splitting into sub-specs.

Project settings

Read project in scripts/pipeline_lanes.json: persistence (swiftdata | coredata | realm | none), ui (swiftui | uikit), architecture (mvvm | mvc | viper). If the key is absent, assume swiftdata, swiftui, mvvm. Apply the persistence and UI rules below for the configured values, not SwiftData and SwiftUI by default. Read "ViewModel" as Controller (mvc) or Presenter (viper).

Architecture Rules (from AGENTS.md/CLAUDE.md — enforce in every spec)

  • Views contain no business logic
  • Domain Services have zero persistence-framework imports (SwiftData, CoreData or RealmSwift) — 100% unit testable without a simulator
  • <type-safety rule from AGENTS.md, e.g. all money values use Decimal, never Double>
  • ViewModels (or Controllers / Presenters) depend on repository protocols, never concrete implementations
  • New models go in <AppName>/Models/, services in <AppName>/Services/
  • New repository protocols go in <AppName>/Repositories/Protocols/, implementations in <AppName>/Repositories/<Persistence>/ (e.g. SwiftData, CoreData)

Branching

Branch spec/<feature-name> off develop. Save spec to docs/superpowers/specs/YYYY-MM-DD-<feature-name>.md and commit. Open PR to develop.

Done when

The user reviews the spec and says it's approved.

Before handing off to /plan, append to .claude/context/decisions.md:

## YYYY-MM-DD — <Feature Name>
**Approaches considered:** <brief list of approaches from step 4>
**Chosen:** <approach name>
**Reason:** <one sentence — the rationale that drove the decision>

Then hand off to the Planner Agent (/plan).

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.