Keeps long sessions cheap. Hands off to a fresh session at a natural break once the context grows large, and requires an explicit non-Fable model on general-purpose subagent dispatches. States the suite-wide model rule: Sonnet by default, Opus for marked tasks, retries and the final review, Haiku for lookups.
Plugin context-budget from marketplace superpowers-extensions in MarcelRoozekrans/superpowers-extensions. Commands read from the repo's manifest — today.
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.
context-budgetUse when dispatching implementer or reviewer subagents, when choosing a model for a subagent, when a context-budget reminder appears in the conversation, or when handing off to a fresh session at a natural break. Holds the suite-wide model rule (Haiku, Sonnet or Opus, never Fable) and the fresh-sess
node "${CLAUDE_PLUGIN_ROOT}/skills/context-budget/scripts/size-check.mjs"
node "${CLAUDE_PLUGIN_ROOT}/skills/context-budget/scripts/dispatch-check.mjs"
Runs a groomed queue of plans without a human present: denies questions nobody will answer, keeps stops from ending the run early, and logs every call made on the user's behalf. Includes ready-check, which checks the work is ready and writes the queue.
2 skills · hooks
Finds shallow modules in the areas the git history changes most, publishes a ranked report with before/after diagrams, lets you pick one candidate and hands it to brainstorming. Includes codebase-design, the shared deep-module vocabulary for designing and deepening interfaces.
2 skills
Waits for CI as an event, fixes red CI, merges a green PR when asked and follows the release. A hook starts the watch whenever a PR is opened.
1 skill · hooks
Records a finished web feature as a paced browser walkthrough video. A storyboard drives a real Chromium on the running app, then ffmpeg builds chapter MP4s with captions and cards, speeds up slow waits, and checks frames before it reports done.
1 skill · 1 MCP server
AI-powered regression testing for web applications. Discovers existing test suites, runs them, then performs functional checks and visual evaluation via Playwright MCP browser automation at desktop, tablet, and mobile viewports. Generates detailed markdown reports with screenshots.
1 skill · 1 MCP server
Rewrites AI prose tics out of text Claude is about to publish, in English and Dutch. Commit messages, PR and issue text, comments, READMEs and docs are checked against a pattern catalogue and published in their plain form, with no report.
1 skill
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
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.