Install any skill in seconds. Free to start, no credit card required.
Get Started Free →[omh] External app actions - email, Slack, Discord, Notion, Linear, Jira, CRM, and similar providers, scoped with auth, payload, confirmation, and result-evidence gates. Use when the user says: connector-operator, connector operator, external app action, external connector action, saas action, api action, send email, email customer.
.claude/skills/rlaope-omh-apps/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 59% | 0% |
This is a Hermes-native connector-operator workflow skill.
connector-operator exists so Hermes users can ask for this workflow in chat and receive a structured, evidence-bounded OMH operating surface instead of ad hoc narration.
Good example:
prepare_connector_operator_card with required context, wrapper actions, and not-evidence boundaries.Bad example:
achievements, workspace-audit, production-audit, automation-blueprint, github-event-ops, buzz, agent-board, gateway-intent-card, +34 more) - schedules, status, health, and ops review.oh-my-hermes or name the adjacent workflow.omh-routing/references/skill-common-rail.md.Use when Hermes should prepare or supervise a provider-backed external app action without claiming connector availability, credentials, API mutation, delivery, or success.
Strong routing signals: connector-operator, connector operator, external app action, external connector action, saas action, api action, send email, email customer, gmail draft, gmail send, create linear ticket, create linear issue, linear ticket, linear issue, update linear, jira ticket, jira issue, create jira issue, open jira ticket, create jira, notion page, update notion, crm update, salesforce update, hubspot update, create calendar event, calendar invite, google calendar, send slack dm, slack dm, discord dm, post to discord, post to slack, discord post, slack post, connector action, 이메일 보내, 이메일 발송, 메일 보내, gmail 초안, linear ticket, linear 티켓, linear 이슈, jira 티켓, jira 이슈, notion 페이지, 노션 페이지, 캘린더 초대, 외부 앱, 외부 커넥터, 커넥터 액션
Category: connector Phase: connector-task Hermes role: guide Quality tier: workflow-surface-gated Reasoning demand: standard
Quality bar:
Handoff policy:
Keep this as Hermes-facing orchestration guidance first. Prepare executor, connector, gateway, or host-runtime handoff only when the user accepts that next step and observed evidence can be recorded.
Required inputs:
Expected outputs:
Artifact expectations:
Safety rules:
Preferred harness for this skill: connector-operator.
shomh runtime record --skill connector-operator --harness connector-operator --status started
Record observed delegation results; otherwise return not_available or not_observed. Prepared OMH routing is not execution, review, CI, merge-readiness, or merge evidence.
Preserve workflow intent and stop conditions; verify before claiming completion.
Use Hermes-native subagent/delegation features when available: native subagents -> Hermes delegation when available, otherwise sequential lanes.
Shared product, compatibility, topology, memory, harness, and execution rules: omh-routing/references/skill-common-rail.md. Load it when applicable; otherwise name an unavailable capability.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-19 | fail→pass | 11,359 | 6,063 | -47% | 1 | 1 | 0% | 1,864 | 2,510 | +35% | 0 | 0 | — |
case-20 | fail→pass | 11,536 | 6,855 | -41% | 1 | 1 | 0% | 2,129 | 2,545 | +20% | 0 | 0 | — |
case-21 | fail→pass | 15,349 | 11,493 | -25% | 1 | 1 | 0% | 2,659 | 3,534 | +33% | 0 | 0 | — |
case-01 | fail→pass | 17,014 | 38,129 | +124% | 1 | 1 | 0% | 3,123 | 2,935 | -6% | 0 | 0 | — |
case-02 | fail→pass | 11,097 | 8,222 | -26% | 1 | 1 | 0% | 1,775 | 2,815 | +59% | 0 | 0 | — |
case-03 | fail→pass | 9,780 | 8,904 | -9% | 1 | 1 | 0% | 1,817 | 3,022 | +66% | 0 | 0 | — |
case-04 | pass→pass | 5,319 | 7,323 | +38% | 1 | 1 | 0% | 689 | 2,652 | +285% | 0 | 0 | — |
case-05 | fail→pass | 17,686 | 9,293 | -47% | 1 | 1 | 0% | 2,559 | 2,975 | +16% | 0 | 0 | — |
case-06 | pass→pass | 5,855 | 7,481 | +28% | 1 | 1 | 0% | 868 | 2,706 | +212% | 0 | 0 | — |
case-07 | fail→pass | 15,563 | 11,682 | -25% | 1 | 1 | 0% | 2,324 | 3,279 | +41% | 0 | 0 | — |
case-08 | fail→pass | 15,085 | 6,932 | -54% | 1 | 1 | 0% | 2,380 | 2,621 | +10% | 0 | 0 | — |
case-09 | fail→pass | 12,699 | 9,294 | -27% | 1 | 1 | 0% | 1,908 | 3,028 | +59% | 0 | 0 | — |
case-10 | fail→pass | 6,958 | 7,416 | +7% | 1 | 1 | 0% | 1,073 | 2,652 | +147% | 0 | 0 | — |
case-11 | fail→pass | 17,737 | 11,363 | -36% | 1 | 1 | 0% | 2,857 | 3,431 | +20% | 0 | 0 | — |
case-12 | fail→pass | 15,527 | 11,567 | -26% | 1 | 1 | 0% | 2,469 | 3,455 | +40% | 0 | 0 | — |
case-13 | fail→pass | 10,850 | 8,081 | -26% | 1 | 1 | 0% | 1,691 | 2,923 | +73% | 0 | 0 | — |
case-14 | fail→pass | 5,391 | 5,071 | -6% | 1 | 1 | 0% | 923 | 2,402 | +160% | 0 | 0 | — |
case-15 | fail→pass | 6,305 | 6,622 | +5% | 1 | 1 | 0% | 883 | 2,553 | +189% | 0 | 0 | — |
case-16 | pass→pass | 9,376 | 9,265 | -1% | 1 | 1 | 0% | 1,673 | 3,003 | +79% | 0 | 0 | — |
case-17 | fail→pass | 19,256 | 10,833 | -44% | 1 | 1 | 0% | 3,093 | 3,379 | +9% | 0 | 0 | — |
case-18 | fail→pass | 15,778 | 7,919 | -50% | 1 | 1 | 0% | 2,693 | 2,774 | +3% | 0 | 0 | — |
case-22 | fail→pass | 14,667 | 7,794 | -47% | 1 | 1 | 0% | 2,554 | 2,795 | +9% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +86 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.