테스트·보안 스킬

Analyze software protections during authorized reverse engineering, the examination of compiled programs. Helps identify mechanisms that hinder debugging, obscure code, or prevent tampering, then choose safe analysis methods, document findings, and develop defensive recommendations.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Analyze protected software with written permission
• Examine malware within an authorized scope
• Study protection mechanisms in CTFs or approved training

Test Android app flows defined in XML by performing taps, swipes, and text entry, then checking screen states. Produce a JSON report with step results, commands, and failure reasons, stopping at the first failure.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Verify login flows against an XML test specification
• Test product search and cart navigation
• Investigate UI failures on devices or emulators

Test app interfaces on an Android emulator using ADB, a command-line device tool. Inspect element positions, perform taps and text entry, then check screenshots and React Native logs to verify behavior and investigate layout or interaction bugs.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Verify app interfaces after UI changes
• Investigate layout and interaction bugs
• Check feature behavior through automated interactions

Compare prompt revisions on a fixed set of test cases to identify regressions, meaning behavior that gets worse after a change. Record model settings, case-level evidence, missing results, costs, and evaluation limits for a reviewable decision.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Comparing original and revised prompts
• Checking required JSON fields after prompt changes
• Reviewing tool selection and task completion

Investigate failed Agent QA runs using logs, recorded artifacts, and local source code. Make a focused code or test-definition repair, then verify the affected behavior while keeping product and infrastructure defects visible.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Repair a failed staging test from recorded evidence
• Update an outdated Agent QA YAML definition
• Distinguish test failures from product or infrastructure defects

Configure permission checks for unattended Antigravity CLI runs. The setup combines blocking rules, automatic approval for scoped operations, AI classification for ambiguous calls, and single-use approval tokens to control tool access within a verified workspace.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Setting up permission checks for unattended Antigravity CLI runs
• Defining rules for shell commands, file changes, and network access
• Configuring a local model to classify ambiguous tool calls

Design layered protections for AI agents against malicious instructions, tool misuse, and data leakage. Provides guidance and code examples for input checks, isolated tool execution, permission limits, and audit logging to help review and strengthen agent security.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Review defenses against malicious instructions
• Isolate tool execution and restrict permissions
• Assess data leakage risks between tenants

Plan structured adversarial tests for AI applications to assess safety bypasses, data leaks, and unauthorized tool use. Organize test prompts, record findings, and prioritize fixes using reporting guidance and an adaptable testing template.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Assess an AI feature before release
• Compare security behavior of candidate models
• Retest after a prompt injection or jailbreak incident

Create and edit Agent QA tests, test suites, and hooks, then validate their structure and tool-generated IDs before running them. Use MCP, an agent tool connection, or the CLI within a configured workspace.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Create and validate a checkout test for staging
• Update an existing test suite
• Check YAML definitions and IDs

Plan reviews of AI-generated code before running or merging it. Define language-specific checks, tests in isolated environments, and evidence requirements, while recording coverage limits and gaps that prevent approval.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning reviews of AI-generated code before merging
• Defining language-specific security checks
• Recording test evidence and unresolved gaps

Define security controls for AI coding agents used by a team. Covers command and file permissions, secret scanning, isolated execution environments, code review gates, and audit trails, with configuration examples to adapt.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Set team rules before introducing AI coding agents
• Restrict access to credentials and infrastructure files
• Design an isolated environment for agent execution

Investigate failed Agent QA runs using recorded steps, logs, and saved evidence. Classify each failure, explain confidence and supporting findings, and identify the likely area to fix and the next action.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Investigating a failed or interrupted Agent QA run
• Reviewing saved evidence and execution logs
• Comparing recurring failures across recent runs

Review GitHub Actions workflows that run AI agents, tracing how external input reaches prompts and tools. Produce findings with evidence, severity, impact, and remediation guidance through static analysis, which examines configuration without running attacks.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Auditing AI agent integrations in GitHub Actions
• Tracing issue and PR text into agent prompts
• Reviewing sandbox settings and tool permissions

Plan security controls for LLM applications and AI agents, covering malicious instructions, data leakage, model tampering, and API access. Includes guidance and code examples for filtering outputs, enforcing permissions, isolating networks, and recording audit events.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing defenses against prompt injection in AI agents
• Filtering personal information from LLM responses
• Designing authentication and rate limits for LLM APIs

Plan periodic access reviews for AWS IAM, GitHub, Okta, and other identity systems. Provides guidance for identifying unused access, assigning reviewers, documenting approval or removal decisions, and organizing evidence for compliance audits.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning periodic access reviews
• Reviewing inactive accounts and unused credentials
• Checking access against current job responsibilities

Test how agent workflows recover from repeatable simulated failures, including tool timeouts, worker restarts, and lost memory. Produces a fault matrix, event timeline, and verdict showing whether state, safety boundaries, and retry budgets were preserved.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Test recovery from MCP or tool timeouts
• Verify checkpoint recovery after a worker restart
• Check outcomes when one parallel branch fails

Compare agent, prompt, or tool configuration changes using versioned test cases and explicit outcome checks. Record each attempt, distinguish agent failures from evaluation or infrastructure problems, and report regressions, permission violations, and uncertainty.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Compare agent behavior before and after prompt changes
• Reproduce an observed failure with preserved evidence
• Run regression tests after a bounded fix

Review code and infrastructure for security risks, map potential threats, and plan stronger defenses. The skill covers OWASP checks, attack and defense exercises, and incident response, with validation required for each environment.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing code and infrastructure security
• Mapping threats with STRIDE or PASTA
• Planning stronger security settings

Check web pages and components for accessibility issues against WCAG 2.2, web accessibility guidelines. Receive a report prioritized by user impact, or request targeted code fixes followed by verification of resolved and newly introduced issues.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Get a prioritized accessibility report for a page
• Audit accessibility patterns across components
• Fix reported issues and verify the changes

Compare accessibility findings between two versions of a web page. Identify newly introduced and fixed issues, count existing findings, and report evidence and suggested fixes without editing application code.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check accessibility before committing local changes
• Compare a page against a branch baseline
• Confirm which accessibility issues a change fixed

Audit a live web page for accessibility issues and produce a fix worklist with exact element selectors, evidence, and suggested changes. Findings include impact counts and source locations when available, without editing the application.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check accessibility after a UI change
• Locate contrast and accessible-name issues
• Prepare a fix worklist with element selectors

Audit web or mobile pages and user journeys against an agreed WCAG accessibility standard. Combine automated scans with manual checks to document barriers, prioritize fixes, and record retest results alongside any remaining gaps.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Audit keyboard access through checkout
• Check screen-reader access, focus, and contrast
• Prioritize accessibility fixes

Support authorized security assessments of Microsoft Active Directory with guidance on environment discovery, credential risks, authentication weaknesses, and privilege escalation. The material organizes tools and workflows for penetration testing and controlled training environments.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning an authorized Active Directory assessment
• Exploring attack paths with BloodHound
• Reviewing authentication and privilege weaknesses

Turns supplied AI agent evaluation results into reports that distinguish autonomous success, human-assisted success, failure, timeout, and invalid runs. Makes metric denominators, retries, timing, costs, and comparison conditions explicit to support evidence-based decisions.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Separating autonomous and human-assisted success rates
• Reporting evaluations with retries and invalid runs
• Comparing agents or releases under matched conditions