A blue-team research agent that builds positive investment arguments from evidence. It reviews an evidence ledger, a record of supporting sources, instead of treating summaries or one press release as strong proof.
A chief-arbitrator research agent that combines opposing investment arguments into a final strategic report. It compares claims with source records and confidence reports, which show how reliable each conclusion is.
A research-planning agent that turns a business question into checkable claims and evidence plans. It can also prepare interview guidance when interviews are part of the research.
A Chinese-language research agent that challenges business claims by looking for missing evidence, copied sources, conflicts, and unsupported promotional statements.
Use when the user asks to analyze a GitHub repository's issues/PRs to identify high-value commercial demand opportunities. Triggers: 「扫issue」「扫PR」「找需求」「分析仓库」「找商业化点子」「独立开发者切入点」「需求雷达」「扫一下明星项目」「生成需求报告」. Outputs scored demand cards with JTBD analysis, RICE scoring, Ugly Workaround evidence, and fake-door probe drafts.…
Use whenever the user asks to monitor AI builder signals, GitHub AI projects, model/app launch opportunities, hot keywords, PMF signals, commercial trend detection, or daily AI radar workflows. Sondar converts early signals into STRIKE/WATCH/IGNORE/KILL decisions with evidence, owner, deadline, metric, and commercial…
Use when the user asks to find true AI builders like a seed person, discover who to follow, monitor what builders are shipping, expand an AI builder watchlist, or separate builder signals from influencer noise.
AGENTS.md instructions for geekjourneyx/imgcli, covering agents, engineering policy, required checks before merge, when to run real smoke and version and release rules.
Use this skill to run imgcli for agent-native image processing tasks: image inspection, normalization and format conversion, fixed-layout card composition, recipe-driven workflow execution, multi-platform variant export, smart padding to platform presets, packaging images into PDFs with visible text watermarks, and…
Research and stress-test open-source project names, from collision checks to launch-ready decisions. Use when the user asks to name or rename a project, compare candidate names, screen GitHub or package namespace risk, verify language or mythology claims, or improve SEO/GEO clarity. Perform live web and registry…
Repository instructions for agent-fs, a project whose documentation covers project memory, design principles, structure, and the steps before committing code. CLAUDE.md is a file that tells a coding agent how to work in a repository.
Agent-first cross-platform file operations CLI. Use when AI Agent needs local file operations (zip/unzip/info/read) or cloud storage operations (upload/download/list) with S3-compatible providers.
Build or upgrade Go CLI projects to production quality with a deterministic workflow: architecture, stable CLI contract, tests/lint gates, cloud smoke tests, release automation, installer, README quality, and skill packaging. Use when users ask to develop a Go CLI from scratch, improve engineering quality, fix…
Use this skill to operate the gcli Gmail CLI for practical email tasks: inbox triage, Gmail query search, message/thread retrieval, full content extraction, and machine-readable output handling. Trigger when users ask to查邮件、筛选邮件、读完整正文、按条件搜索、提取附件相关邮件、或直接执行 gcli 命令并解释结果.
Use this skill whenever the user wants to migrate from OpenClaw to Hermes Agent, upgrade an OpenClaw workspace into Hermes, preserve skills/memory/cron/messaging/provider settings, debug Feishu or Weixin gateway issues after migration, or validate that Hermes is ready to replace OpenClaw. This skill is especially…
A post-production tool for turning existing animation HTML or video files into deliverable videos, GIFs, and music versions. It handles exporting, audio mixing, batch variants, and checks of the resulting files.