dandye/ai-runbooks/full-alert-triage
Claude Code and agent skill by dandye/ai-runbooks
- community
Checking that the SKILL.md file is still in the repository.
Complete Tier 1 triage workflow. Orchestrates the full alert triage process: check-duplicates, triage-alert, enrich-ioc for each entity, and either close (FP/BTP) or escalate (TP/Suspicious). Use for end-to-end alert processing.
Install
npx vskill@latest install dandye/ai-runbooks/full-alert-triageSKILL.md
Full Alert Triage Workflow
A composite skill that orchestrates the complete Tier 1 alert triage process from initial receipt to disposition (close or escalate).
Inputs
CASE_IDorALERT_ID- The alert/case to triage (required)
Orchestrated Workflow
Detailed Steps
Phase 1: Pre-Check
Step 1.1: Check for Duplicates
Invoke: /check-duplicates CASE_ID=$CASE_ID
- If duplicate confirmed:
- Invoke:
/document-in-casewith "Closing as duplicate of [Similar Case ID]" - Invoke:
/close-case-artifactwith reason NOT_MALICIOUS - END WORKFLOW
- If not duplicate: Continue to Phase 2
Phase 2: Initial Triage
Step 2.1: Perform Alert Triage
Invoke: /triage-alert CASE_ID=$CASE_ID
Extract from results:
CLASSIFICATION- FP, BTP, TP, or SuspiciousKEY_ENTITIES- List of IOCs (IPs, domains, hashes, URLs)ALERT_TYPE- Type of alert (malware, authentication, network, etc.)PRIORITY- Suggested priority level
Phase 3: Enrichment
Step 3.1: Enrich Each Entity
For each entity in KEY_ENTITIES:
Invoke: /enrich-ioc IOC_VALUE=$entity
Collect:
GTI_FINDINGS- Threat intelligence resultsSIEM_CONTEXT- SIEM entity summaryIOC_MATCH_STATUS- Whether IOC appears in threat feeds
Update CLASSIFICATION if enrichment reveals new information.
Phase 4: Decision & Action
Step 4.1: Make Final Classification
Based on triage and enrichment, confirm classification:
| Classification | Criteria | Action | |
|---|---|---|---|
| False Positive (FP) | No malicious indicators, known benign | Close | |
| Benign True Positive (BTP) | Real but authorized/expected | Close | |
| True Positive (TP) | Confirmed malicious | Escalate | |
| Suspicious | Inconclusive, warrants investigation | Escalate |
Step 4.2: Execute Disposition
If FP or BTP:
- Invoke:
/document-in-casewith:
- Classification and rationale
- Evidence summary from enrichment
- Closure justification
- Invoke:
/close-case-artifactwith:
- Reason: NOT_MALICIOUS
- Root cause: Appropriate option (e.g., "Legit action", "Normal behavior")
If TP or Suspicious:
- Invoke:
/document-in-casewith:
- Classification and rationale
- Evidence summary
- Recommended next steps
- Output escalation recommendation:
- Escalate to Tier 2
- Suggest appropriate follow-up skill based on alert type:
- Malware →
/triage-malware - Authentication →
/triage-suspicious-login - IOC-focused →
/deep-dive-ioc
Phase 5: Report
Step 5.1: Generate Triage Report
Invoke: /generate-report REPORT_TYPE=triage
Include:
- Case/Alert ID
- Classification with rationale
- Key entities and enrichment results
- SIEM queries executed
- Disposition taken
- Next steps (if escalated)
Outputs
| Output | Description | |
|---|---|---|
FINAL_CLASSIFICATION | FP, BTP, TP, or Suspicious | |
DISPOSITION | Closed or Escalated | |
EVIDENCE_SUMMARY | Key findings from triage and enrichment | |
REPORT_PATH | Path to generated triage report | |
ESCALATION_TARGET | If escalated, recommended next skill/tier |
Error Handling
- If
/check-duplicatesfails → Log warning, continue with triage - If
/enrich-iocfails for an entity → Log warning, continue with other entities - If
/close-case-artifactfails → Log error, manual closure required - If any MCP tool unavailable → Document limitation, proceed with available data
Performance Targets
- Total workflow time: < 15 minutes
- Duplicate detection: < 1 minute
- Per-entity enrichment: < 2 minutes
- Target accuracy: > 90% correct classification
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationWarning, 70 of 100Marginal. 1 run. Evaluated 8 months ago.
Source
- SKILL.md in the repositoryCheckingskills/_workflows/full-alert-triage/SKILL.md
Popularity
- Repository stars
- 90
- Repository forks
- 25
- 7-day trend score
- 23
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 24 Mar 2026, 7 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/dandye/ai-runbooks/full-alert-triage)