Test design, vulnerability assessment, and quality assurance (13.1K skills)
Develop behavior through one failing test and one minimal implementation at a time. Tests use agreed interfaces and expected results from specifications or documented examples, producing evidence of both failure and success, including for bug fixes.
Compare a UI screenshot with reference images and receive a structured JSON assessment. Get a score, pass or revision status, concrete visual differences, and suggested edits for layout, spacing, typography, colors, and visual hierarchy.
Write and register SGLang unit and end-to-end tests, from choosing models and file locations to reusing server fixtures. Follow project rules for meaningful assertions, CI registration, dependency mocking, and safe server cleanup after tests.
Build and debug sglang tests comparing token probabilities during prompt processing and step-by-step generation. Separate numerical differences from reused-state bugs, choose test thresholds, and locate the first operation that diverges.
Guide test design for SGLang's scripted runtime by deciding when a test harness API adds real value. Favor direct reads, meaningful outcome checks, and the engine's actual execution loop over unnecessary wrappers.
Investigate recurring SGLang automated test failures by comparing logs, commit history, and GPU environments. Narrow down code changes or machine-specific problems, optionally reproduce failures remotely, and summarize the evidence in a structured report.
Review existing tests and CI, the automated checks run during development, for redundant coverage, misplaced tests, and opportunities to reduce runtime or cost. Receive file-specific recommendations based on documented improvement patterns, with uncertain findings separated.
Set up Zeek to observe network traffic and produce structured logs for security analysis. Create custom detection scripts, investigate suspicious DNS activity and repeated external connections, and forward logs to a central security monitoring system.
Analyze ModSecurity, AWS WAF, and Cloudflare logs for SQL injection, attempts to manipulate database queries through web requests. Groups related requests, tracks source IPs, classifies attack methods, and produces incident reports.
Investigate suspicious privilege increases, including Windows UAC approval bypasses and Linux sudo or setuid abuse. Use registry changes, process relationships, and permission changes to identify possible abuse and document affected hosts, methods, and risk levels.
Find npm and PyPI packages that imitate trusted package names to exploit typing mistakes. Compare spelling similarity, publication dates, downloads, and maintainer details, then rank suspicious candidates and produce reports or blocklists for further review.
Use CloudFox to inspect AWS and Azure resources through read-only queries. Identify exposed endpoints, secrets, permissions, and role trust relationships, then organize findings into files for reviewing configuration weaknesses and potential paths to broader access.
Evaluate threat intelligence platforms, which collect and organize cyber threat information. Define requirements, compare integrations and operating costs, plan practical trials, and use weighted scoring to select a platform and outline its rollout.
Configure Fail2ban to detect port scans, repeated SSH login attempts, and suspicious web requests from server logs. Create custom detection patterns, automatically block source IP addresses, and set up notifications and longer bans for repeat offenders.
Investigate Windows credential theft by correlating endpoint monitoring data, Sysmon events, and security logs. The skill provides detection queries and a reporting format to identify suspicious access and assess which credentials may be compromised.
Find WMI event subscriptions that let attackers retain access to Windows systems. Correlate Sysmon events with stored subscriptions and Autoruns results, inspect execution triggers and commands, and document risks and recommended remediation.
Use Stratus Red Team to simulate attack techniques across AWS, Azure, GCP, and Kubernetes. Prepare test resources, run and repeat simulations, check logs and alerts, and remove resources to assess whether security detections work.
Build real-time detection for ransomware encryption by combining changes in data randomness with file modification rates, renames, and process activity. Define behavioral scores and response thresholds, then check detection accuracy and monitoring overhead.
Assess whether a container can escape its isolation and reach the host during an authorized security test. Review privileges, shared mounts, Docker socket exposure, and runC vulnerabilities, then document demonstrated impact and remediation.
Find undocumented API endpoints, the addresses applications send requests to, by comparing traffic with OpenAPI specifications and examining cloud settings and source code. Build an inventory with activity and risk details to review forgotten or outdated APIs.
Assess an authorized Active Directory environment by simulating attacks against credentials, permissions, and trust relationships. Use BloodHound and related testing tools to identify paths to administrator access, evaluate detection capabilities, and document findings with remediation recommendations.
Configure email security gateways to detect targeted phishing messages. Set up sender impersonation protection, link checks, isolated attachment analysis, and custom rules, then validate quarantine actions and alerts for suspected attacks.
Assess serverless functions, primarily AWS Lambda, for unsafe execution of external input and unauthorized changes to code or added libraries. Combine source inspection, activity logs, and permission reviews to identify injection risks and recommend detection and prevention measures.
Screen dependency names on npm, PyPI, and crates.io before installation to find lookalikes of legitimate packages. This skill compares spelling patterns and registry metadata, helping review AI-generated lists, lockfile changes, and new dependencies in CI/CD.