Decompose a whole application spec into a dependency-ordered graph of RED-able vertical slices — the planning step above frame-checks. Cut the app into the smallest observable outcomes (a user can sign up, a link redirects), wire the edges from setup dependencies (B can't be tested until A's behavior exists), route cross-cutting concerns (auth, audit, quotas, notifications) and heavy shared engines (pipelines, scorers) as inherited obligations/infrastructure rather than nodes, seat one or more walking skeletons so no node is scaffold-only, name the invariants every later slice must preserve, and emit only the currently-buildable frontier — then re-plan as each slice ships and new requirements surface. The output is a build plan (an emergent DAG + invariants) that feeds frame-checks one frontier at a time and promotes verified loops into a batch manifest. Use when someone has a multi-feature app/product spec and wants to build it as a loop-of-loops, asks "how do we break this spec into loops / a batch," or needs a build order before authoring any .loop.yaml.
Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI
Supported by the Atlas install prompt. Skill-specific requirements vary.
Skill stats
Skills1CategoryDeveloper toolsStars0
About this skill
Decompose a whole application spec into a dependency-ordered graph of RED-able vertical slices — the planning step above frame-checks. Cut the app into the smallest observable outcomes (a user can sign up, a link redirects), wire the edges from setup dependencies (B can't be tested until A's behavior exists), route cross-cutting concerns (auth, audit, quotas, notifications) and heavy shared engines (pipelines, scorers) as inherited obligations/infrastructure rather than nodes, seat one or more walking skeletons so no node is scaffold-only, name the invariants every later slice must preserve, and emit only the currently-buildable frontier — then re-plan as each slice ships and new requirements surface. The output is a build plan (an emergent DAG + invariants) that feeds frame-checks one frontier at a time and promotes verified loops into a batch manifest. Use when someone has a multi-feature app/product spec and wants to build it as a loop-of-loops, asks "how do we break this spec into loops / a batch," or needs a build order before authoring any .loop.yaml.
Quick start
Copy the install prompt. Use the green button above this README.
Send it to your agent. Paste it into your coding agent’s chat and send.
Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Decompose a whole application spec into a dependency-ordered graph of RED-able vertical slices — the planning step above frame-checks. Cut the app into the smallest observable outcomes (a user can sign up, a link redirects), wire the edges from setup dependencies (B can't be tested until A's behavior exists), route cross-cutting concerns (auth, audit, quotas, notifications) and heavy shared engines (pipelines, scorers) as inherited obligations/infrastructure rather than nodes, seat one or more walking skeletons so no node is scaffold-only, name the invariants every later slice must preserve, and emit only the currently-buildable frontier — then re-plan as each slice ships and new requirements surface. The output is a build plan (an emergent DAG + invariants) that feeds frame-checks one frontier at a time and promotes verified loops into a batch manifest. Use when someone has a multi-feature app/product spec and wants to build it as a loop-of-loops, asks "how do we break this spec into loops / a batch," or needs a build order before authoring any .loop.yaml.
About this listing
Category
Developer tools
Reported license
Not specified
Listing language (estimated)
English
Review status
Not individually reviewed
Agent compatibility
Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLI