superpowers
claudecodexCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Plan software work in a Schematic Planner project: design canvases in specs, executable task canvases in plans, shared by you and the agent.
Plugin schematic-planner from marketplace schematic-planner in ridanit-ruma/schematic-planner-plugin. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
brainstorming-on-canvasUse after project discovery to select or create a specification canvas in specs and design a requested change before implementation planning.
connectUse when a user wants to connect Schematic Planner or its canvas tools are missing because the MCP server is unconfigured or has no credential.
connecting-the-canvasUse when the Schematic Planner tools are missing, refused, or unconfigured — no list_plans/create_plan in the tool list, an authorization error from one of them, or the person asking to connect an agent to their instance.
dispatching-parallel-agentsUse when an approved Schematic Planner implementation Plan has at least two ready tasks that may be independent; scan for conflicts before bounded parallel delegation and fall back to sequential execution when overlap remains.
executing-plans-on-canvasUse after implementation Plan approval to carry out one ready task from a Plan in the bound project's plans folder, keeping the canvas honest as you go.
finishing-a-development-branchUse after all tasks in an approved implementation Plan are complete; review the whole branch, verify it freshly, and offer only explicitly authorized keep, PR, merge, or delete actions.
receiving-code-reviewUse when review findings arrive for a Plan task; verify each finding against the approved Spec, current code, and runnable evidence before fixing, rejecting, or requesting material clarification.
requesting-code-reviewUse after a Plan task has fresh implementation evidence and before accepting it as complete; request Spec-compliance review first, then code-quality review, and record both outcomes on the Plan.
subagent-driven-developmentUse when an approved Schematic Planner implementation Plan has a bounded ready task and the runtime and user allow delegation; assign one task, then have the coordinator verify and record the result.
systematic-debuggingUse when a test, command, or runtime behavior fails unexpectedly or persists after an attempted fix; reproduce it, trace the shared path, test one causal hypothesis, and fix the root cause.
test-driven-developmentUse when implementing a non-trivial approved Plan task whose behavior can be checked before and after the change; observe the relevant failure, make the smallest fix, and preserve fresh evidence.
using-git-worktreesUse before executing an approved Plan task when concurrent work or isolation may materially reduce risk; inspect repository state and optionally create a safe Git worktree without disturbing user changes.
using-schematic-plannerUse when starting any conversation in a repository planned on Schematic Planner; bind the repository to a project, discover its specs and plans, and route work before anything is built.
verification-before-completionUse immediately before claiming work is fixed, passing, or complete or marking a Plan task done; run fresh checks that directly prove the claim and report their exact results and limits.
writing-plans-on-canvasUse after Spec approval to create or update a separate executable implementation Plan in the bound project's plans folder.
writing-skillsUse when a recurring Schematic Planner plugin behavior gap may require creating or substantially revising a workflow skill; prove the gap, prefer extending its current owner, and validate behavior as well as structure.
/schematic-planner:connectPoint every agent on this machine at a Schematic Planner instance
"${CLAUDE_PLUGIN_ROOT}/hooks/run-hook.cmd" session-start
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.
3 skills · 6 commands · 1 MCP server · hooks
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
8 skills · 1 MCP server
Use Alibaba Open Code Review from local Codex.
2 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.