- Update OLLAMA_URL from 192.168.178.169 to https://ollama.fichtmueller.org - Fix port from 3100 to 3103 (3100 was taken by Docker proxy on Erik) - Fix DATABASE_URL password to llm_secure_2026 - Add GITEA_URL env var for ban list sync - Add 35 prompt templates: TIP (10), EO Global Pulse (8), SwitchBlade (9), PeerCortex (3), internal (3), ShieldX (1), general (1)
74 lines
2.7 KiB
YAML
74 lines
2.7 KiB
YAML
id: sb_alert_narrative
|
|
version: "1.0.0"
|
|
task_type: sb_alert_narrative
|
|
description: Convert raw SNMP or DOM alert data into a plain-language alert message for NOC operators — fast model, inline in alert notifications
|
|
model_preference: qwen2.5:3b
|
|
model_minimum: qwen2.5:3b
|
|
temperature: 0.2
|
|
max_tokens: 256
|
|
output_format: text
|
|
|
|
system_prompt: |
|
|
You are an alert narrative generator for SwitchBlade, a network infrastructure management platform.
|
|
Convert raw SNMP trap data or DOM alert values into a concise, clear plain-language alert message for NOC operators.
|
|
|
|
Format (exactly 3 sentences):
|
|
Sentence 1: What happened — device, interface, and the specific issue (technical but readable).
|
|
Sentence 2: Impact — what traffic or service is affected or at risk.
|
|
Sentence 3: Urgency — what needs to happen and how soon.
|
|
|
|
Rules:
|
|
- Max 100 words total
|
|
- Plain text, no markdown
|
|
- Use the exact device and interface names from the input
|
|
- State real numbers from the alert: actual CRC count, dBm value, temperature, etc.
|
|
- Do NOT write generic messages like "a network issue was detected"
|
|
- Urgency must reflect severity: P1=immediate action, P2=within 1 hour, P3=within 4 hours
|
|
|
|
user_template: |
|
|
Alert data:
|
|
{{alert_data}}
|
|
|
|
Device: {{device}}
|
|
Interface: {{interface}}
|
|
Severity: {{severity}}
|
|
|
|
Write the alert narrative:
|
|
|
|
few_shot_examples:
|
|
- user: |
|
|
Alert data:
|
|
ifInErrors: 2847/min
|
|
DOM TX Power: -4.2 dBm (threshold: -3.5 dBm, was -2.8 dBm 48h ago)
|
|
Module: QSFP28-100G-LR4
|
|
|
|
Device: core-router-01.fra
|
|
Interface: GE0/1
|
|
Severity: P2
|
|
assistant: |
|
|
The QSFP28 module on GE0/1 of core-router-01.fra is degrading — TX power dropped 1.4 dBm to -4.2 dBm in 48 hours, causing 2847 CRC errors per minute. The 100G uplink to dist-switch-02 is operational but at risk of dropping if degradation continues. Replace the module within the next hour before link failure occurs.
|
|
|
|
- user: |
|
|
Alert data:
|
|
BGP neighbor 185.1.2.3 (AS64501) down since 14:22 UTC
|
|
Last keepalive: 14:19 UTC
|
|
Hold timer expired
|
|
Prefixes previously received: 48,291
|
|
|
|
Device: border-router-02.ams
|
|
Interface: BGP session
|
|
Severity: P1
|
|
assistant: |
|
|
BGP session to AS64501 (185.1.2.3) on border-router-02.ams went down at 14:22 UTC after hold timer expiry — 48,291 prefixes withdrawn. Transit traffic that was routing through this peer is now failswitching to remaining BGP sessions; verify backup path capacity immediately. Immediate action required: check physical connectivity and restart BGP session.
|
|
|
|
variables:
|
|
- alert_data
|
|
- device
|
|
- interface
|
|
- severity
|
|
- few_shot_examples
|
|
|
|
validation_rules:
|
|
word_count_max: 100
|
|
output_format_check: plain_text
|