Masih-0x3
- 150 skills
- 0 followers
- last month last updated
- ▌ Diffui Designer · masih-0x3 bundleRuns the diffui MCP workflow before any HTML/CSS/JS is written: create_project, generate_options (4 variants), generate_page, create_build_link, then implement code from the build markdown. Auth (whoami/authenticate) only on tool errors — installs usually include DIFFUI_API_KEY already. REQUIRED when the user asks to create, build, or design a webpage, website, landing page, app screen, dashboard, or UI — including "create a webpage for …", "design first then build/create in …", themed sites (jungle, arctic, etc.), and output paths like demoBuild/. Do NOT skip diffui and design manually in prose or jump straight to coding when the diffui MCP server is available. Also use when the user mentions diffui.
- ▌ Domain Modeling · masih-0x3 bundleUse when project terminology, product workflow, data contracts, or architectural decisions are causing ambiguity, or when a feature/RCA/refactor needs a shared domain vocabulary. In local Codex work, read existing repo docs first and write durable docs only when the repo convention or user request justifies it.
- ▌ Frontend Design · masih-0x3 bundleCreate or improve production-quality frontend UI/UX in code, including React, Next.js, Tailwind CSS, shadcn/Radix, responsive layouts, design systems, accessibility, motion, landing pages, dashboards, visual redesigns, and frontend QA. Use when the user asks for UI, UX, frontend design, visual polish, web or app screens, components, responsive or mobile fixes, landing pages, design-to-code work, or screenshot-backed frontend verification. When Magic UI components, animations, backgrounds, or marketing UI could help, use the globally configured Magic UI MCP.
- ▌ Marketing Ideas · masih-0x3 bundleWhen the user needs marketing ideas, inspiration, or strategies for their SaaS or software product. Also use when the user asks for 'marketing ideas,' 'growth ideas,' 'how to market,' 'marketing strategies,' 'marketing tactics,' 'ways to promote,' 'ideas to grow,' 'what else can I try,' 'I don't know how to market this,' 'brainstorm marketing,' or 'what marketing should I do.' Use this as a starting point whenever someone is stuck or looking for inspiration on how to grow. For specific channel execution, see the relevant skill (ads, social, emails, etc.).
- ▌ Pick UI Library · masih-0x3Use when the user asks which frontend library or package to choose for UI components, motion, charts, interaction, state, styling, or related infrastructure.
- ▌ Steel Developer · masih-0x3 bundleUse this skill when the user wants reusable code, scripts, examples, or docs that build on Steel cloud browsers with SDKs, REST APIs, Playwright, Puppeteer, Stagehand, Browser Use, credentials, profiles, files, extensions, embeds, proxies, or CAPTCHA APIs. Do not use for live web browsing performed by the agent; use steel-browser. Route failed-session diagnosis to steel-session-debugging and reliability mitigation to steel-reliability.
- ▌ Toolcraft Figma · masih-0x3Toolcraft: use when a generated app task includes a Figma URL, node ID, or design-to-code requirement.
- ▌ Transitions Dev · masih-0x3 bundleUse when implementing or replacing CSS transitions for badges, dropdowns, modals, panels, page changes, resizing cards, number or text swaps, icon swaps, success and error states, hover effects, search clearing, skeletons, shimmer, tabs, tooltips, staggered reveals, 3D tilt, menu morphs, accordions, or expand and collapse behavior.
- ▌ Gitea Github Cicd · masih-0x3 bundleValidate repository changes through a server-hosted Gitea Actions route and then push the identical verified commit to GitHub. Use when a user asks to set up, repair, run, migrate, or operate CI/CD with Gitea on codex-linode, Gitea Actions, Gitea Runner, GitHub remotes, or a Gitea-before-GitHub quality gate.
- ▌ Review Animations · masih-0x3 bundleReviews animation and motion code against a high craft bar derived from Emil Kowalski's design engineering philosophy. Default to flagging; approval is earned.
- ▌ Steel Reliability · masih-0x3 bundleUse this skill when a Steel automation appears unreliable because of bot detection, CAPTCHA loops, 403/access-denied or verify-you-are-human pages, proxy failures, profile or identity problems, credential or login persistence issues, suspicious pacing, redirect loops, or retry strategy, or when the user asks about stealth, proxies, CAPTCHA solving, profiles, credentials, or reliable logins. Do not use as a first step for unknown failed sessions; collect evidence with steel-session-debugging first when needed.
- ▌ Toolcraft Browser · masih-0x3Toolcraft: use to verify generated apps in a real local browser after implementation.
- ▌ Verified Research · masih-0x3 bundleUse when the user asks to research, investigate, compare sources, verify claims, evaluate evidence, or produce an evidence-backed answer or dossier without immediately implementing changes.
- ▌ Audit Orchestrator · masih-0x3 bundleOrchestrate a scoped, goal-backed multi-agent audit loop from one parent thread only when the audit is large or complex enough to benefit from delegation. Use when the user asks for an audit conductor/orchestrator, scoped audit plan, UI/UX/backend/security/performance/product audit with subagents, or wants an audit plan turned into /goal and iterated until done criteria are satisfied. Do not use for small one-file reviews, narrow bug checks, or implementation-first work; use a direct audit/review instead.
- ▌ Cline Orchestrator · masih-0x3 bundleDelegate supervised UI/UX design, visual implementation, design audits, image-to-code, plan critiques, or other benchmark-approved slices from Codex to Cline/ClinePass/GLM through the terminal. Use when Codex should own and steward Cline as a child specialist: keep Cline current, inspect Cline skills, select Cline only when current benchmark evidence says it helps, run GLM 5.2 with xhigh reasoning, capture provenance, recover interrupted Cline work through Codex, incorporate only reviewed output, and verify the final project behavior.
- ▌ Obscura Browser MCP · masih-0x3 bundleUse the local Obscura MCP browser server from Codex as the default worker browser for fast DOM/text web automation, page extraction, form interaction, cookies/storage checks, console/network diagnostics, and agent browser work where screenshots are not required. Use when the user mentions Obscura, asks whether an agent can browse or act on their behalf through MCP, wants to verify Obscura MCP access, compare Obscura with Chrome/Playwright, or run web research, scraping, smoke checks, or browser workflows that can be proven with DOM/text evidence rather than visual screenshots.
- ▌ Reverse Engineering · masih-0x3 bundleProvides reverse engineering techniques. Use when the main job is to understand how a compiled, obfuscated, packed, or virtualized target works before exploiting or solving it, including binaries, APKs, WASM, firmware, custom VMs, bytecode, malware-like loaders, and anti-debug or anti-analysis logic. Do not use it when the vulnerability is already understood and the remaining task is exploitation; use pwn instead. Do not use it for pure web workflows, log or disk forensics, or standalone crypto problems unless reversing the implementation is the real blocker.
- ▌ Steel Skill Creator · masih-0x3 bundleUse this skill when the user wants to turn a recurring browser workflow into a reusable, parameterized agent skill, especially when the task has concrete inputs and a clear output such as scheduled scrapes, form submissions, data extraction, monitoring flows, price probes, or login-gated reports. Do not use for one-off web tasks; use steel-browser.
- ▌ Pr CI Status Monitor · masih-0x3 bundleCheck pull requests, CI failures, review comments, CodeRabbit feedback, GitHub issue/bounty status, Algora or payout claim state, and required next actions. Use when the user asks to check PR status, CI status, failing checks, review comments, CodeRabbit comments, whether something was merged/paid/rewarded/blocked, contribution or bounty status, or asks for a concise monitor/update on GitHub work.
- ▌ Writing Great Skills · masih-0x3 bundleUse when creating, editing, reviewing, or adapting agent skills in ~/.agents/skills or ~/.codex/skills, especially to reduce ambient bloat, decide model-vs-user invocation, tighten descriptions, prune duplicate instructions, or make a skill fit the local Codex operating style.
- ▌ Codex Chatgpt Control · masih-0x3Use when Codex agents need to operate visible ChatGPT web sessions through the codex-chatgpt-control SDK, including prompts, threads, files, downloads, reports, browser bridge blockers, and local source smokes.
- ▌ Coding Agent Sessions · masih-0x3Use when asked to inspect, reconstruct, audit, compare, or summarize prior coding-agent work across Codex, Claude, OpenCode, or local session logs, including token usage, subagents, tool calls, archived rollouts, disconnected work, or questions about what happened since a specified time.
- ▌ Planning Orchestrator · masih-0x3 bundleOrchestrate research and save an implementation-ready plan file through a goal-backed planning loop for large product, UI/UX, architecture, migration, or implementation programs. Use when the user asks for a planning conductor/orchestrator, comprehensive revamp plan, /goal-backed plan, design plan from reference images, website/product inspiration research, multi-surface project plan, or a roadmap to hand to an implementation orchestrator. Do not use for small tasks, one-file changes, audits of existing defects, or when the user wants code edits immediately.
- ▌ Safari Web Extensions · masih-0x3 bundleBuild, package, test, and reason about Safari Web Extensions for iOS, iPadOS, and macOS. Use when creating or modifying Safari extensions, WebExtension manifests/content scripts/background scripts/popups/options pages, extension-native app messaging, App Store packaging, Safari permissions, or Safari-specific debugging.
- ▌ Software Orchestrator · masih-0x3 bundleUse when invoking Software Orchestrator as a /goal-style run with an attached plan/PRD/doc until every in-scope item is verified. Includes cold-start priors for host models (Cline-pass, Antigravity, Kilo free): strengths, thinking levels, specialty routing. Grok orchestrates; workers execute; live outcomes update the capability store.
- ▌ Framer Code Components · masih-0x3Framer code component implementation guidance, platform constraints, layout annotations, property controls, and authoring best practices. Use only after loading the framer skill and reading the Components row of the generated projects/safeProjectId/index.md task map first in the same task; never load this skill directly as the entry point.
- ▌ Checkpoint Quality Loop · masih-0x3 bundleUse when the user wants a serious checkpoint before continuing work, especially after meaningful implementation, refactor, audit remediation, launch-readiness, or multi-agent work where correctness, production quality, maintainability, or a scoped lens like UI/UX, security, backend, data, performance, architecture, or release readiness must be verified.
- ▌ Project Task Decomposer · masih-0x3 bundleUse when converting a PRD, product handoff, spec, architecture brief, roadmap, or implementation plan into a versioned, dependency-aware task corpus (hundreds to thousands of leaf tasks) for a Software Orchestrator. Supports PRD_ONLY, PRD_PLUS_REPO, and CORPUS_UPDATE. Does not implement the product or permanently assign models.
- ▌ Release Handoff Summary · masih-0x3 bundleSummarize latest changes, release readiness, deployment state, verification status, changelog notes, documentation updates, and remaining next steps. Use when the user asks "what changed", "latest changes", "what is left", "next tasks", "release prep", "changelog", "handoff", "status update", "are we ready to ship", "summarize this project", or wants a concise done/left/blocked plan after implementation.
- ▌ Root Cause Investigator · masih-0x3 bundleRoot cause analysis for bugs, failures, regressions, incidents, broken flows, failing tests/builds/CI, production errors, or odd behavior. Use when the user says something is broken, failing, disappearing, timing out, returning errors, not loading, not working, behaving oddly, asks for root cause analysis/RCA, says "don't change anything yet", or wants diagnosis of UI, backend, auth, database, API, webhooks, logs, deploys, third-party integrations, CI, or production behavior. Start read-only, use logs/telemetry when available, narrow plausible causes to the proven cause, then produce a minimal fix plan.
- ▌ Steel Session Debugging · masih-0x3 bundleUse this skill when the user provides a failed Steel session ID, failed automation run, timeout, browser error, blocked page, or unexpected result and wants evidence-backed diagnosis from session metadata, browser logs, raw agent logs, semantic traces, replay links, screenshots, network failures, or errors. Do not use for first-pass bot mitigation; collect evidence here, then hand off to steel-reliability for proxy, CAPTCHA, identity, or anti-bot fixes.
- ▌ Toolcraft Brainstorming · masih-0x3Toolcraft: use before creating generated app features, changing behavior, or assembling app specs.
- ▌ Toolcraft Writing Plans · masih-0x3Toolcraft: use before code changes once generated app behavior or an approved spec is clear.
- ▌ Src Hunter · masih-0x3 bundle实战 SRC / 众测 / Bug bounty 漏洞挖掘工作流 skill。包含:5 阶段方法论(intake → recon → enum → hunt → report)、19 个攻击类 playbook(SQLi/XSS/RCE/SSRF/IDOR/CSRF/Path Traversal/File Upload/SSTI/XXE/Race/HTTP Smuggling/OAuth/JWT/SAML/GraphQL/Mobile/LLM/DoS)、305 个结构化 payload、263 个 WAF/EDR 绕过变体、2887 份 HackerOne 真实 High/Critical 已披露案例、77,000+ WooYun 案例统计、国产 OA / 中间件指纹库、银行 / 电信行业垂直 playbook。当用户提到 "src 挖洞 / src 漏洞挖掘 / bug bounty / 众测 / hackerone / 漏洞赏金 / SRC / 任意 X 漏洞 / 渗透测试" 或问"如何挖某个目标 / 怎么测某个 API / 如何绕过 WAF" 时触发。
- ▌ Resume Disconnected Work · masih-0x3 bundleResume disconnected, stale, or out-of-context Codex work from available local evidence. Use when the user says to resume, continue previous work, recover a lost thread, check whether prior work was done, or continue after a disconnect. Do not use for a normal fresh task with a clear current repo and no prior-state dependency.
- ▌ Unknown Diff Commit Gate · masih-0x3 bundleReview an ambiguous dirty git worktree before committing. Use when the user says changes may or may not be worthy, asks to inspect uncommitted work, commit only good changes, separate user/concurrent edits, or decide whether local diffs are safe. Do not use when the user already specified an exact commit scope.
- ▌ Production Readiness Gate · masih-0x3 bundleAudit launch, deploy, staging, production, release, or go-live readiness from evidence. Use when asked whether an app/site/workflow is ready to ship, public-launch ready, deployment-ready, production-ready, or what blocks release. Do not use for ordinary code review unless the user needs a readiness verdict.
- ▌ Repo Audit To Remediation · masih-0x3 bundleAudit a local repository, identify the highest-value defects or gaps, implement focused fixes, verify them, and leave a concise backlog or handoff. Use when the user asks for a codebase audit, comprehensive review, top priority issues, improvements, remediation plan, "fix the top issues", security/UX/performance/test/docs audit, repo health check, or an audit-to-patches workflow.
- ▌ Workflow Mining To Skills · masih-0x3 bundleMine accessible Codex, ChatGPT, repo, memory, transcript, or project history to identify repeated workflows and create or update skills. Use when asked to review prior threads/sessions, workflow-mine, author skills from repeated work, deduplicate skills, or decide what belongs in skills versus AGENTS/scripts/automations. Do not use when creating a single user-specified skill without history analysis.
- ▌ Redesign Existing Projects · masih-0x3 bundleOn-demand audit and targeted upgrade of an existing website or app's design.
- ▌ Reviewing Capability Usage · masih-0x3 bundleUse when the user asks to review recent skill, tool, subagent, browser, memory, connector, goal, or local capability usage, especially to find missed capabilities, overuse, overclaiming, redundancy, or whether a new skill is justified.
- ▌ Background Browser Operator · masih-0x3 bundleWork with browser surfaces in a controlled background lane while the user studies, meets, reviews, or does another primary task. Use Obscura MCP as the default browser labor surface for DOM/text workflows, and escalate to Chrome, in-app browser, or computer-use for visual, authenticated, profile-specific, extension, or browser-parity proof. Use when the user asks to keep browser work running in the background, monitor or verify a web flow without interrupting them, open a target before manual handoff, or pair browser verification with planning, audit, implementation, production-readiness, Safari extension, or frontend-design work.
- ▌ Implementation Orchestrator · masih-0x3 bundleOrchestrate goal-backed implementation for large plans, audit remediation, RCA fix plans, multi-phase features, UI/UX revamps, backend/data/security work, or cross-surface engineering changes. Use when the user asks for an implementation conductor/orchestrator, says to implement a saved planning-orchestrator file, audit report, or root-cause fix plan, or wants focused subagents/workers coordinated until acceptance criteria pass. Do not use for small one-file fixes, read-only audits, planning-only work, or unresolved failures that still need root cause analysis first.
- ▌ X Opportunity Reality Check · masih-0x3 bundleFact-check and evaluate X posts, GitHub repos, trading bots, Polymarket/Kalshi/crypto/AI automation claims, side-income ideas, bounty claims, and business opportunities for realism. Use when the user asks "is this real", "is this legit", "can this make money", "how realistic is this", sends an X post/thread/article with a repo or guide, asks about trading bots, delta-neutral or arbitrage claims, passive income claims, GitHub project viability, or wants odds, risks, and a practical go/no-go recommendation.
- ▌ Find Animation Opportunities · masih-0x3Use when the user asks what could be animated or wants an interface to feel more alive; inspect code, reject unnecessary motion, and report precise read-only suggestions without implementing them.
- ▌ Engineering Acceptance Review · masih-0x3 bundleUse when the user wants completed Codex or agent work checked before accepting it: project-goal fit, task fit, hallucinated code, spaghetti, architecture drift, tests, browser behavior, security/data risk, evidence, and final accept/fix/block verdict. Use after meaningful generated code, repo edits, UI changes, refactors, bug fixes, audit remediation, or implementation plans.
- ▌ Toolcraft Systematic Debugging · masih-0x3Toolcraft: use before fixing broken controls, tests, builds, visual mismatches, exports, or runtime regressions.
- ▌ Toolcraft Figma Implement Design · masih-0x3Toolcraft: use to translate inspected Figma structure into a production-ready generated app.
- ▌ Research Plan Implementation Loop · masih-0x3 bundleUse when the user asks for research that should lead to planning, implementation, product changes, build decisions, or an end-to-end loop rather than a standalone research answer.
- ▌ Dsl Vm Reverse · masih-0x3Reverse custom JavaScript or WASM virtual machines and risk engines. Use when a target contains an interpreter loop, encoded opcodes, constant tables, or a custom DSL runtime.