Identify software protections that hinder inspection, such as debugger detection and obfuscation that makes code harder to understand. Support authorized binary analysis and CTF security challenges with protection findings, bypass strategies, reports, and analysis scripts.
- Identify protections in an authorized malware sample
- Plan analysis of a packed executable
- Implement debugger detection for a CTF challenge
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 4, 2026
Turn a one-sentence agent goal into a routed work phase and an execution plan, such as a single run, a capped grade-and-improve loop, or a scheduled cron loop. The skill also guides setup questions and handoff to phase-specific skills.
- Planning a Claude Managed Agent from a clear task goal
- Choosing between one-time, iterative, and scheduled runs
- Routing agent setup to the appropriate work phase
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 4, 2026
Review an AI Management System against ISO 42001 before certification, an internal audit, or onboarding an AI system. The skill checks scope, policy, risk coverage, reassessment after model changes, audit planning, and reuse of existing management system evidence.
- Preparing for an ISO 42001 stage 1 audit
- Checking AI system coverage in an AIMS scope
- Planning an internal audit and reviewing auditor independence
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 4, 2026
Design or audit native interfaces for Apple platforms against the Human Interface Guidelines. Check measurable details such as text contrast and tap target size with a bundled tool, then report violations and manual review items.
- Planning a new iOS or macOS interface
- Checking contrast and tap target sizes in a mockup
- Reviewing an Apple platform UI before release
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 4, 2026
Review recorded agent sessions to find what changed, inspect events in order, and check whether claims such as “tests passed” are supported. It separates recorded facts from rule-based inferences and points answers to specific trace steps.
- Investigating why an agent changed a file
- Checking which files a past run touched
- Verifying whether tests passed after the last edit
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 4, 2026
Compares HR rosters with account exports and checks access review campaigns for missing coverage, self-approval, and incomplete revocations. It maps findings to controls across common security and compliance frameworks, producing evidence for review.
- Finding enabled accounts belonging to former employees
- Checking access review coverage and unresolved revocations
- Preparing evidence for an auditor’s access-control request
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 3, 2026
Use this skill to assess an AI system against key EU AI Act requirements before EU deployment or during an intake review. It checks risk classification, conformity routes, company roles, transparency duties, and obligations for general-purpose AI, with Article citations and next steps.
- Screening an AI system before EU deployment
- Checking whether an AI use may be prohibited or high-risk
- Planning conformity assessment for a high-risk system
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 3, 2026
Use this skill to decide how much authority an AI agent should have in marketing and go-to-market workflows. It sorts decisions by reversibility, exposure, and frequency into AI-primary, collaborative, or human-primary zones, then defines review and approval rules for each.
- Setting review rules for AI-generated marketing work
- Deciding whether an agent can publish or only draft
- Auditing review requirements across many workflows
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 3, 2026
Plan and carry out an AngularJS to Angular migration, from choosing a rewrite strategy to converting components, services, and routes. Includes a hybrid approach that runs both frameworks together while individual features are migrated.
- Planning an AngularJS to Angular migration
- Moving a large app one feature at a time
- Converting directives into Angular components
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 3, 2026
Use this skill to turn a goal into a task plan with explicit checks, then run tasks one at a time through a controller that verifies evidence, limits retries, and escalates when needed. It supports resumable work through durable plan and state files.
- Breaking a broad goal into verifiable tasks
- Running a multi-step task with bounded retries
- Resuming goal work from saved plan and state files
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 3, 2026
Create and verify a signed, tamper-evident receipt for consequential agent actions such as deployments, payments, deletions, or access grants. The receipt records the agent, operation, target, and governing policy, and can be verified offline from the certificate alone.
- Create evidence for an agent deployment or deletion
- Record a payment or access grant under a policy
- Verify a signed action receipt offline
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 2, 2026
Audit React, Next.js, Vue, Angular, Svelte, or HTML projects against WCAG 2.2 A and AA. Find accessibility issues, check color contrast, suggest framework-specific fixes, verify changes, and prepare compliance reports or CI results.
- Checking a web project for WCAG 2.2 issues
- Finding contrast problems in CSS or Tailwind colors
- Reviewing keyboard access and ARIA attributes
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 2, 2026
Run several AI agents on the same task in separate Git worktrees, then compare their results using a metric or quality review. Merge the selected approach for work such as code optimization, refactoring, bug fixes, research, or copy variations.
- Comparing parallel approaches to a coding task
- Benchmarking optimization attempts
- Generating and judging copy variations
LeoYeAI/agent-master-skills☆ 2.2KUpdated Nov 2, 2026