superpowers
claudecodexCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
DeepStack raises a task in altitude, depth, and range, then proves the result. Principles follow Felix Rieseberg, How to be more ambitious: A tutorial for humans with models, in this plugin's own words.
Plugin deepstack from marketplace DeepStack in jl-cmd-projects/DeepStack. Commands read from the repo's manifest — 19 days 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.
deepstack-helpExplain DeepStack, /deepstack-mode, and which playbook fits a task. Use for /deepstack-help. A help question does not start the work.
deepstack-modeRaise a task in altitude, depth, and range, then prove it. Use for DeepStack, /deepstack-mode, or a first task.
principle-assemble-the-crewApply when one field is too narrow for the verbs. Pick the fields whose methods the job needs.
principle-challenge-from-raw-dataApply when the task treats a belief as given. Reproduce that belief from raw material.
principle-durable-loopApply when the outcome should continue with no person feeding the next task. Build the access first.
principle-methods-leave-impersonation-outApply when a draft starts speaking as a character. Ask for the method that field runs.
principle-nouns-to-verbsApply when the task is named as an artifact. Restate it as the action the artifact is for.
principle-outcome-above-the-taskApply when the task is one guess at a larger outcome. Name what should happen next.
principle-outside-judgeApply before a done claim. The worker does not grade its own result.
principle-produce-new-knowledgeApply once the worker is inside the raw material. Ask for a blind spot that predicts on unseen material.
principle-room-to-missApply when fear of missing is shrinking the attempt. Give the attempt room to fail.
principle-same-instrumentsApply when the draft changes wording and leaves the method alone. Require the tools the best people in the field use.
principle-seen-successApply when a goal looks unreachable and one case of it already exists. Point at that case.
principle-specific-hard-goalsApply when the ask is a vague best effort. Set a specific hard goal.
principle-speed-is-settledApply when a request is only finishing work already planned, and only faster. DeepStack leaves that kind alone.
principle-unreasonable-and-solvableApply when the target is a mild improvement. Set a specific hard target and treat it as solvable.
setup-deepstackChoose models per DeepStack role and turn the session hook on or off. Use for /setup-deepstack.
deepstack-agentDelegate for a DeepStack run. Use for a helper. The parent chat runs deepstack-mode. Read skills/deepstack-mode/SKILL.md before any work. A fresh agent takes each new task.
"${CLAUDE_PLUGIN_ROOT}/hooks/session-start.sh" claude
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Scaffold, validate, and publish new Claude Code plugins with proper structure
3 skills · 1 command · 1 agent
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
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
8 skills · 1 MCP server
Perform AI code review on Git diffs — supports workspace changes, branch ranges, and single commits with concurrent per-file analysis, codebase search, and deep context-aware review.
2 commands
Claude Code skill for creating and presenting Slidev slidedecks
1 skill
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.