Testing skills

This skill guides turning bug reports into mergeable fixes by following minimal reproduction, root-cause hypotheses, smallest change, tests, and gate

This skill guides you through Gandalf prompt-injection challenges with a structured, educational approach to safely practice red-teaming techniques.

This skill audits Flatnotes and Tasks.md against GitHub as truth, surfaces drift, and plans fixes to keep docs and projects aligned.

This skill orchestrates AI-driven usability tests with Nova Act, generating personas, collecting data, interpreting results, and producing HTML reports.

This skill verifies completion of multi-step tasks against the original plan, ensuring quality criteria are met and documenting outcomes.

This skill helps architects design and implement SwiftUI/UIKit architectures with MVVM, MVI, TCA, or VIPER guidelines tailored to your project.

This skill helps AI agents stay honest and grounded by applying practical drift checks, ego awareness, and uncertainty practices in every session.

This skill performs PubChem queries and RDKit analysis to retrieve compound data, properties, structures, and plan retrosynthesis routes.

This skill generates consistent Mermaid diagrams across templates, validating content and rendering outputs for technical visuals.

This skill analyzes Solidity contracts for security vulnerabilities, suggests fixes, and generates audit reports for Hardhat and Foundry projects.

This skill sets up authorized C2 simulation workflows and measures defensive detection outcomes for secure, compliant pentest operations.

This skill pre-validates ClawHub skills by scanning actual content with Snyk and blocking installs that contain prompts, secrets, or malware.

This skill helps you separate speakers, compare voices, and process audio for cloning and verification using Demucs, pyannote, and Resemblyzer.

This skill helps you create and enforce natural language spending policies for Sardis agent wallets, enabling safe budgeting and automated approvals.

This skill validates Volcengine TOS AK/SK configuration by listing buckets and performing upload, download, and pre-signing checks to ensure access works.

This skill performs multi-agent adversarial security reviews by enlisting three LLMs to debate findings and expose real vulnerabilities.

This skill guides reviewing task-linked PRs end-to-end, from context gathering and diff review to post-merge handoff and Trello updates.

This skill autonomously designs, implements, tests, and delivers complete projects end-to-end from plain-English prompts for rapid review and deployment by you.

This skill helps you operate the deepbook CLI end-to-end for market data, config, and on-chain trades across spot, swap, and margin.

This skill evaluates mathematical expressions safely and returns accurate results while protecting against unsafe code execution and edge cases.

This skill helps you manage Volcengine TOS storage tasks efficiently by performing bucket and object operations with validation and logging.