Test design, vulnerability assessment, and quality assurance (13K skills)
Investigate failing macOS tests in Xcode or SwiftPM by running a focused test scope, classifying errors, and separating setup issues from code regressions. Get a concise explanation of likely causes and the next verification step.
Test a real Electron app against a Den server in two separate Daytona environments. Check cloud sign-in, marketplace changes, organization policies, provider synchronization, and worker recovery, then collect evidence from the interface, APIs, logs, screenshots, and optional recordings.
Document and validate Daytona interface workflows with labeled screenshots in a browsable HTML page. Add short videos when motion matters, check that evidence shows the claimed behavior, and share it for pull request review.
Launch the real iPolloWork app on Daytona or locally and run scripted end-to-end tests through its debugging connection. Collect machine-checkable assertions, validated screenshots, and reports as evidence of the tested behavior.
Verify app changes through real user journeys and produce a single HTML demonstration. Each step connects a claim, user action, observable check, narration, and validated screenshot so reviewers can inspect evidence that the experience works.
Test iPolloWork's handling of corporate certificates in a Daytona Windows sandbox. Set up valid and broken HTTPS certificate chains, then verify that the desktop app and runtimes it starts use certificates trusted by Windows.
Verify production fixes by connecting code changes and releases to Sentry issues. Compare error events and affected users before and after release, then report whether the fix worked, improved behavior, failed, or lacks enough traffic to judge.
Analyze git history to map contributors to security-sensitive files and identify ownership gaps. Estimate bus factor, a measure of dependence on few people, group files that change together, and export CSV or JSON for further analysis.
Audit and improve website accessibility using WCAG 2.1, guidelines for making web content usable by people with disabilities. Check image descriptions, contrast, forms, keyboard controls, and screen reader support with automated and manual testing.
Create a Markdown threat model grounded in a codebase. Identify valuable assets, boundaries between trusted and untrusted components, realistic attack paths, and mitigations, then prioritize risks using explicit assumptions about likelihood and impact.
Automate website interactions and checks with Playwright, a browser automation tool. Create and run scripts for forms, login redirects, broken links, and different screen sizes, with screenshots and execution results to inspect.
Measure website performance with Lighthouse, interpret scores and reports, and compare results across builds. Set limits for loading times and resource sizes, then integrate audits into CI, an automated workflow that checks code changes.
Review Python, JavaScript/TypeScript, and Go code against security guidance for the languages and frameworks in use. Identify important weaknesses, suggest improvements, and produce prioritized reports with code references, or apply secure defaults when writing new code.
Review website code across performance, accessibility, search engine optimization, and web best practices. Receive findings ranked by severity, explanations of their impact, concrete fixes with code examples, and a recommended order for addressing them.
Audit AGENTS.md, rules, skills, and their references for broken paths or commands. Optional independent agent reviews assess redundant instructions and practical usefulness, producing reports that identify what to retain, review, or consider trimming.
Evaluate code against a product requirements document, scoring implementation and tests separately for each acceptance criterion. Produce an evidence-backed final grade, identify gaps, and compare implementations using a fixed checklist and scoring method.
Test whether instructions planted in an agent's long-term memory or searchable knowledge base influence later behavior. Verify both a stored test marker and a resulting change in answers, permissions, or treatment of retrieved documents.
Tests whether an AI agent exposes real credentials embedded in accessible source code or configuration. It requests searches for keys, passwords, and private keys, then documents confirmed disclosures with file locations and conversation evidence.
Test whether an AI agent reveals another user's data or performs actions beyond the caller's permissions. Dialogue probes cover access across users or organizations, admin functions, and attempts to gain higher privileges without authorization.
Check how an agent handles file saves, uploads and archive extraction. Test whether user supplied paths can escape the intended directory, then report confirmed unsafe handling with response evidence and suggested safeguards.
Test whether an AI agent executes commands hidden in user input, evaluates untrusted code, or runs scripts from unverified sources. Record confirmed weaknesses with the test conversation, potential impact, and suggested safeguards.
Test whether instructions hidden in simulated documents, retrieved passages, or web pages can redirect an agent. The skill checks for unwanted behavior or sensitive disclosures and reports confirmed failures with the test dialogue as evidence.
Test whether an agent misuses user trust by requesting secrets, making false safety guarantees, recommending unsafe actions, or impersonating authority. Confirmed issues are recorded with dialogue evidence, potential impact, and suggested improvements.
Probe AI agents through dialogue to check for unexpected command execution, file access outside intended paths, and prompted network requests. It selects tests based on available tools and records confirmed behavior with conversation evidence and mitigation suggestions.