4,053 tools found
A prompt-layer ruleset that forces LLM agents through a hypothesis-verify-yield loop before executing state-changing commands.
Docket turns an AI coding agent's transcript into a per-commit evidence record, so code review attention goes where nothing was verified.
agent-dev-team routes coding tasks to 21 tiered AI role agents, escalating instead of letting one assistant attempt everything.
Astah Pro MCP lets AI agents design, explain, and generate code from UML diagrams inside Astah Professional.
Design statistically rigorous A/B and multivariate tests, calculate sample size, and run an ICE-prioritized growth experimentation program.
Claude skill that runs coding tasks as a state machine, escalating to a human instead of claiming done without evidence-backed acceptance criteria.
Claude skill that audits WCAG 2.2 accessibility issues via live-DOM inspection and either reports them or fixes and verifies them.
Accesslint Diff parses command-line arguments to extract branch names and URLs for accessibility linting workflows, enabling branch-specific or default-branch
Accesslint Scan audits a live web page and reports selector-grounded WCAG violations without editing any code.
Claude skill that adds an iOS App Clip target to an Expo app, wiring associated domains, AASA hosting, and TestFlight submission.
ADHX retrieves clean JSON data from any X/Twitter post or long-form article via a free API, optimized for LLM consumption without scraping or browser
Production-grade LLM evaluation skill implementing direct scoring, pairwise comparison, rubric generation, and bias mitigation for automated quality assessment.
