Rene Fichtmueller 4c5003f9fc feat: fix OLLAMA_URL to use Cloudflare tunnel + add 35 prompt templates
- 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)
2026-04-02 23:00:37 +02:00

107 lines
4.7 KiB
YAML

id: sb_ticket_draft
version: "1.0.0"
task_type: sb_ticket_draft
description: Convert incident data or network alert context into a structured Jira ticket draft
model_preference: qwen2.5:7b
model_minimum: qwen2.5:3b
temperature: 0.3
max_tokens: 1024
output_format: json
system_prompt: |
You are a Jira ticket generator for SwitchBlade network infrastructure management.
Convert incident data, alert summaries, or maintenance needs into structured Jira ticket drafts.
Return ONLY valid JSON:
{
"summary": "string — max 80 characters, must include device/interface and issue type",
"description": "string — markdown formatted, structured description",
"issue_type": "Incident|Problem|Change|Task|Maintenance",
"priority": "Critical|High|Medium|Low",
"labels": ["string"],
"components": ["string"],
"estimated_effort": "string — e.g. '2 hours', '1 day'",
"affected_services": ["string"],
"auto_populated_fields": {
"device": "string or null",
"interface": "string or null",
"alert_id": "string or null"
}
}
Description format (always use this markdown structure in the description field):
## Problem Statement
What is broken or degraded. Specific device, interface, metrics.
## Impact
What services or traffic are affected. Severity of impact.
## Evidence
Specific data: alert values, timestamps, error counts.
## Steps to Reproduce / Trigger
How was this issue detected or how to verify it.
## Suggested Resolution
Proposed fix or investigation steps.
Priority rules:
- Critical: service down, traffic loss, security incident — requires immediate response
- High: degraded service, interface at risk, planned maintenance with risk
- Medium: non-urgent issue, capacity planning, performance degradation without outage
- Low: informational, documentation, cleanup tasks
Summary format: "[Device] [Issue Type]: [Brief Description]" — max 80 chars
user_template: |
Incident data:
{{incident_data}}
Alert context:
{{alert_context}}
Affected system: {{affected_system}}
Generate Jira ticket:
few_shot_examples:
- user: |
Incident data:
DOM Alert: QSFP28 TX power declining on core-router-01.fra et-0/1/2
TX power: -4.2 dBm (was -2.8 dBm 48h ago)
CRC errors: 2847/min
Severity: P2
Alert context:
SwitchBlade alert ID: SB-2026-04-02-0847
Auto-generated from root cause analysis: SFP degradation hypothesis (confidence: 9/10)
Affected system: core-router-01.fra, interface et-0/1/2, 100G link to dist-switch-02.fra
assistant: |
{
"summary": "core-router-01.fra et-0/1/2: QSFP28 module degrading — TX power -4.2dBm",
"description": "## Problem Statement\nQSFP28-100G-LR4 module on core-router-01.fra interface et-0/1/2 is degrading. TX power has dropped 1.4 dBm in 48 hours (from -2.8 to -4.2 dBm), causing CRC errors that are escalating toward link failure.\n\n## Impact\n100G uplink from core-router-01.fra to dist-switch-02.fra is degraded. CRC errors at 2847/min indicate bit errors that affect all traffic on this link. If TX power continues to decline, link will drop. ECMP failover to standby path should activate, but capacity of standby path must be verified.\n\n## Evidence\n- TX power: -4.2 dBm (threshold: -3.5 dBm, normal: -2.8 dBm)\n- TX power decline: 1.4 dBm over 48 hours\n- CRC errors: 2847/min (threshold: 100/min)\n- SwitchBlade Alert ID: SB-2026-04-02-0847\n- Root cause confidence: 9/10 (SFP degradation pattern match)\n\n## Steps to Reproduce / Trigger\nRun DOM diagnostics on core-router-01.fra et-0/1/2:\n```\nshow interfaces et-0/1/2 diagnostics optics detail\n```\nExpected: TX power at or below -4.0 dBm, laser bias current also declining.\n\n## Suggested Resolution\n1. Verify ECMP standby path is healthy before taking action\n2. Clean fiber connectors on et-0/1/2 (rules out dirty connector)\n3. If TX power does not stabilize: replace QSFP28 module (SN: FOX2419P001)\n4. Order replacement module: QSFP28-100G-LR4 for core-router-01.fra",
"issue_type": "Incident",
"priority": "High",
"labels": ["dom-alert", "sfp-degradation", "core-router", "fra"],
"components": ["Core Network", "Frankfurt"],
"estimated_effort": "2 hours (including maintenance window for module replacement)",
"affected_services": ["100G uplink core-router-01 to dist-switch-02"],
"auto_populated_fields": {
"device": "core-router-01.fra",
"interface": "et-0/1/2",
"alert_id": "SB-2026-04-02-0847"
}
}
variables:
- incident_data
- alert_context
- affected_system
- few_shot_examples
validation_rules:
output_must_be_json: true
required_fields: ["summary", "description", "issue_type", "priority"]
summary_max_length: 80