Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Set up Telegram, Discord, or Slack webhooks for engineering layer alerts — long-running task completion, build failures, security audit alerts.
.claude/skills/evolution-foundation-dev-configure-notifications/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -13% | 0% |
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
Configure outbound notifications for engineering layer events: long-running task completion, build failures, critical security findings, etc.
dev-autopilot finishes (it can take minutes)@vault-security to ping you on CRITICAL findingsint-telegram integrationCommon triggers:
dev-autopilot completion@vault-security CRITICAL finding@oath-verifier FAIL verdict@hawk-debugger 3-failure circuit breakerint-telegram)Save webhook config to .claude/settings.json or env vars:
json{ "engineeringLayer": { "notifications": { "telegram": { "enabled": true, "chatId": "${USER_TELEGRAM_CHAT_ID}", "triggers": ["autopilot.complete", "vault.critical", "oath.fail"] }, "discord": { "enabled": false } } } }
Send a test notification via the configured channel. Verify it arrives.
Save config notes to workspace/development/research/[C]notifications-config-{date}.md.
Notification message format:
🤖 [EvoNexus Eng] {agent} {event}
{summary}
📁 {path to artifact}int-telegram (for Telegram delivery)dev-autopilot (most common notification source)@vault-security (CRITICAL finding alerts)update-config (built-in skill for settings edits)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 16,618 | 5,537 | -67% | 1 | 1 | 0% | 2,792 | 1,566 | -44% | 0 | 0 | — |
case-02 | fail→pass | 16,669 | 6,491 | -61% | 1 | 1 | 0% | 3,112 | 1,768 | -43% | 0 | 0 | — |
case-03 | fail→pass | 12,190 | 8,774 | -28% | 1 | 1 | 0% | 2,275 | 2,139 | -6% | 0 | 0 | — |
case-04 | fail→pass | 12,329 | 2,329 | -81% | 1 | 1 | 0% | 2,079 | 945 | -55% | 0 | 0 | — |
case-05 | fail→pass | 5,344 | 1,657 | -69% | 1 | 1 | 0% | 829 | 725 | -13% | 0 | 0 | — |
case-06 | fail→pass | 23,174 | 1,910 | -92% | 1 | 1 | 0% | 1,578 | 807 | -49% | 0 | 0 | — |
case-07 | fail→pass | 14,600 | 3,733 | -74% | 1 | 1 | 0% | 2,666 | 1,191 | -55% | 0 | 0 | — |
case-08 | fail→pass | 11,503 | 2,222 | -81% | 1 | 1 | 0% | 1,813 | 870 | -52% | 0 | 0 | — |
case-09 | fail→pass | 5,911 | 1,858 | -69% | 1 | 1 | 0% | 837 | 796 | -5% | 0 | 0 | — |
case-10 | fail→pass | 10,899 | 3,027 | -72% | 1 | 1 | 0% | 1,806 | 768 | -57% | 0 | 0 | — |
case-11 | fail→pass | 12,906 | 1,825 | -86% | 1 | 1 | 0% | 1,940 | 804 | -59% | 0 | 0 | — |
case-12 | fail→pass | 2,403 | 1,639 | -32% | 1 | 1 | 0% | 347 | 731 | +111% | 0 | 0 | — |
case-13 | fail→pass | 9,117 | 1,232 | -86% | 1 | 1 | 0% | 1,377 | 699 | -49% | 0 | 0 | — |
case-14 | pass→pass | 9,163 | 1,488 | -84% | 1 | 1 | 0% | 1,472 | 720 | -51% | 0 | 0 | — |
case-15 | fail→pass | 8,713 | 2,936 | -66% | 1 | 1 | 0% | 1,421 | 1,009 | -29% | 0 | 0 | — |
case-16 | fail→pass | 8,357 | 1,081 | -87% | 1 | 1 | 0% | 1,286 | 662 | -49% | 0 | 0 | — |
case-17 | fail→pass | 13,481 | 2,011 | -85% | 1 | 1 | 0% | 2,046 | 806 | -61% | 0 | 0 | — |
case-18 | fail→pass | 8,916 | 2,200 | -75% | 1 | 1 | 0% | 1,428 | 873 | -39% | 0 | 0 | — |
case-19 | pass→pass | 7,089 | 5,118 | -28% | 1 | 1 | 0% | 1,257 | 1,464 | +16% | 0 | 0 | — |
case-20 | pass→pass | 10,856 | 6,741 | -38% | 1 | 1 | 0% | 1,942 | 1,705 | -12% | 0 | 0 | — |
case-21 | pass→pass | 12,449 | 13,324 | +7% | 1 | 1 | 0% | 2,187 | 3,061 | +40% | 0 | 0 | — |
case-22 | fail→pass | 10,133 | 2,919 | -71% | 1 | 1 | 0% | 1,973 | 1,142 | -42% | 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, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +82 percentage points is the difference between those two pass rates over the 21 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.