rlaope Oh My Hermes - Setup Guide¶
Source: rlaope/oh-my-hermes (3,000⭐, pushed Sep 28, 2026)
Skill: rlaope/oh-my-hermes (130 installable skills, 1,515+ combined installs on skills.sh)
Category: Agent Workflow Packages / Hermes Operations
Quality Tier: 🔵 Community (new repo, no security verdicts published - verified Sep 28, 2026)
rlaope's Oh My Hermes is an all-in-one plugin for Hermes Agent: "the coding intelligence, a long-term memory system and model optimized workflow packages". Unlike the earlier-documented 9-skill witt3rd/oh-my-hermes suite, this is a much larger (130-skill) collection of Hermes-native workflow packages spanning operations, product, research, sales, finance, security, and the ulw-* work-loop family. Repo pushed the same day as this sweep - the author is actively shipping.
Installation¶
# Individual skills
npx skills add rlaope/oh-my-hermes --skill omh-ops-review
# Or tap the repo as a Hermes skill tap
hermes skills tap add rlaope/oh-my-hermes
hermes skills install omh-ops-review omh-research-brief omh-security-safety-review
Skill Families¶
| Family | Skills | Focus |
|---|---|---|
| omh-ops | 15+ | Ops reviews, observability cards, operating rhythm, run efficiency, workspace audit, morning brief |
| omh-research | 10+ | Research briefs, research department, web research, paper learning, source finder |
| omh-product | 10+ | Product briefs, product discovery/validation, CTO loop, idea-to-deploy, decision prototypes |
| omh-sales | 5+ | Sales development, sales pipeline review, meeting briefs, people ops |
| omh-security | 6+ | Security safety review, security event response, legal compliance, threat models |
| omh-memory | 5+ | Long-term memory system: memory-new, memory-sync, decision recall, instinct ledger |
| omh-code | 15+ | Code review, frontend/backend, refactor plans, build-failure triage, codebase onboarding |
| omh-runtime | 10+ | Toolbelt/connector/executor readiness, prompt import, model setup, routing, meta-router |
| ulw-* | 9 | Work-loop suite: ulw-context, ulw-interview, ulw-plan, ulw-loop, ulw-perf, ulw-qa, ulw-research, ulw-work, ulw-maestro |
| sweeps | 2 | triage-sweep, review-sweep (issue/review batch sweeps) |
Roster - Top Skills¶
| Skill | Installs | Does |
|---|---|---|
| triage-sweep | 19 | Batch triage sweep across an issue/work backlog |
| review-sweep | 19 | Batch review sweep (PRs, documents) |
| ulw-research | 16 | Work-loop research stage |
| omh-ops-review | 16 | Structured ops review of a system or team |
| omh-frontend | 16 | Frontend development workflow |
| omh-decide | 16 | Decision-making workflow with options scoring |
| omh-codegraph-refresh | 16 | Refresh the codebase knowledge graph |
| omh-capability-toggle | 16 | Capability toggle management |
| omh-agent-board | 16 | Agent task board and status tracking |
| omh-ai-slop-cleaner | 15 | Strip AI-slop patterns from writing |
| omh-security-safety-review | 15 | Pre-execution security/safety review gate (prompt injection, secrets, destructive actions) |
| omh-automation-blueprint | 15 | Recurring job blueprint: schedule, delivery, silence policy, context chain |
| omh-browser | 15 | Browser task policy overlay with auth/confirmation/observed-trace gates |
| omh-memory-sync | 15 | Long-term memory sync between sessions |
(130 skills total - 100 indexed on skills.sh at sweep time, installs concentrated at 15-19 each.)
CorpusIQ Use Cases¶
| Use Case | How |
|---|---|
| Cron discipline | omh-automation-blueprint matches the CorpusIQ cron blueprint pattern: schedule + delivery + silence policy |
| Pre-execution gates | omh-security-safety-review aligns with the pre-flight gate doctrine for external actions |
| Weekly self-optimization | omh-ops-review + omh-run-efficiency feed the Monday performance audit |
| Content quality | omh-ai-slop-cleaner pairs with the content-slop-scoring gate for UGC drafts |
| Memory operations | omh-memory-sync / omh-decision-recall complement Honcho + GBrain session handoff |
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
Limitations / Verification¶
- New and fast-moving repo (pushed Sep 28, 2026); skills change frequently - pin versions where possible
- Not the same publisher as the documented witt3rd/oh-my-hermes suite (255⭐, 9 skills)
- Verify:
npx skills add rlaope/oh-my-hermes --listshows the current skill count
Security¶
No skills.sh security audits published (verified Sep 28, 2026):
| Audit | Verdict |
|---|---|
| Gen Agent Trust Hub | Not published |
| Socket | Not published |
| Snyk | Not published |
Related¶
- Oh My Hermes (OMH) Suite Setup (different publisher: witt3rd)
- Oh My Hermes Workflow Setup (different publisher: reason-machines)
- Skills Marketplace
← Skills Catalog | Skills Marketplace → Powered by CorpusIQ