oh-my-agent-skills

Use when importing, installing, packaging, or publishing third-party Hermes skills so category placement, support files, discoverability, and open-source release surfaces stay correct.

Build log monitoring systems that avoid duplicate alerts and false positives. Use when creating cron-based monitoring, log error tracking, or alert deduplication.

Use when creating or refining any Hermes cron job whose final delivery goes to humans on Telegram or similar chat platforms. Prioritize user-readable alert design, compact default output, severity gating, and exception-driven expansion over exhaustive internal dumps.

Use before claiming work is complete, fixed, or passing. Requires fresh verification evidence before any success claim.

Set up or document a Karpathy-style LLM Wiki workflow in Obsidian, especially when raw material should live in Inbox and compiled knowledge should live in _wiki.

Compile one or more raw Obsidian Inbox notes into the `_wiki/` layer using controlled source pages, selective downstream updates, and mandatory index/log maintenance.

Review the health of an existing `_wiki/` compiled knowledge layer, surface priority fixes, and triage backlog candidates. Use after initial wiki scaffolding or periodically during maintenance.

Use when Hermes should run a complete multi-role subagent workflow: decompose work, assign isolated planner/investigator/implementer/reviewer tasks, integrate results, and verify before claiming success.

Default to subagents for all non-trivial work. Use this skill to decide when to delegate, how to partition work for parallel execution, and how to package bulletproof context to prevent subagent drift.

Route URL reading requests across Hermes tools and platform-specific skills. Use when a user shares a URL or asks to read, summarize, extract, inspect, or diagnose content from a webpage, article, post, document link, or similar web resource, especially when a simple fetch may fail or the best tool path is unclear.

Hermes skill optimizer inspired by Karpathy's autoresearch. Evaluates SKILL.md files using an 8-dimension rubric (structure + effectiveness), runs hill-climbing with git version control, validates improvements through test prompts. Use when user mentions "优化skill", "skill评分", "自动优化", "auto optimize", "skill质量检查", "帮我改改skill", "skill怎么样", "提升skill质量", "skill review", "skill打分", "达尔文", "darwin".

Use when creating, revising, or validating Hermes skills so they are discoverable, reusable, and grounded in real successful workflows.