Skip to content
Back to skills

Widefield Security Setup

ASecurity

"Use when the user asks to onboard WideField Security, plan identity threat detection and response,

  • 39 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 9, 2026
securitypythongobashrailsapisecurity

Works with

  • cli
  • api

Security analysis

A100/100

Pro scans all 7 files and shows the line behind each finding

Scanned September 9, 2026

npx -y skills add chambear2809/splunk-cisco-skills --skill widefield-security-setup --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Widefield Security Setup?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Widefield Security Setup
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/chambear2809-widefield-security-setup/badge)](https://www.skillsdirectory.com/skills/chambear2809-widefield-security-setup)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: widefield-security-setup
description: "Use when the user asks to onboard WideField Security, plan identity threat detection and response,
  connect WideField to identity/SIEM/SOAR/governance tools, or coordinate WideField child skill execution
  without using undocumented WideField APIs. Render, route, validate, and optionally delegate a WideField
  Security adoption workflow across Okta, Saviynt, Splunk SIEM, Google SecOps, and identity-threat doctor
  skills."
compatibility: "Splunk Cloud Platform 10.5.2605: delegated. Compatibility is determined by the canonical replacement or selected child skill; this compatibility alias or router does not own a runtime or package."
metadata:
  splunk_cloud_10_5: "delegated"
  compatibility_verified: "2026-08-20"
---

# WideField Security Setup

## Prerequisites

| Tool or access | Purpose | Verify |
|---|---|---|
| Bash and Python 3 | Run bundled setup and validation helpers | `bash --version && python3 --version` |
| Required product/platform access | Inspect or configure the selected target | Complete the documented preflight |
| Credential files for live modes | Keep secrets out of chat | Verify paths only |

## Workflow Overview

```text
┌───────────┐   ┌───────────────┐   ┌───────────────┐   ┌─────────────────┐
│ Preflight │ → │ Render/review │ → │ Apply/handoff │ → │ Validate evidence │
└───────────┘   └───────────────┘   └───────────────┘   └─────────────────┘
```

## When to Activate

- Onboard WideField Security, plan identity threat detection and response, connect WideField to
  identity/SIEM/SOAR/governance tools, or coordinate WideField child skill execution without using undocumented
  WideField APIs.
- Preview and review the widefield security setup workflow before any live apply phase.
- Diagnose failed prerequisites, generated assets, configuration, or validation evidence.

## Scope

Follow the documented read-only or render-first path whenever it is available.
This skill does not imply permission to mutate live systems. Require explicit
apply flags, protected credentials, and operator review for state changes.

## Examples

Inspect the supported setup modes before selecting one:

```bash
bash skills/widefield-security-setup/scripts/setup.sh --help
```

Expected output: usage, supported modes, and required arguments are displayed
without changing the target environment.

Inspect validation modes before running completion checks:

```bash
bash skills/widefield-security-setup/scripts/validate.sh --help
```

Expected output: offline, live, and completion options are displayed when the
skill supports them; help exits without mutation.

## Troubleshooting

| Issue | Cause | Resolution |
|---|---|---|
| Preflight fails | A required tool or access path is missing | Resolve it before rendering or applying |
| Rendered assets are incomplete | Required non-secret inputs are absent | Complete intake and render again |
| Apply is blocked | Review, credentials, or explicit acceptance is missing | Use the documented handoff |
| Validation is incomplete | Live evidence is unavailable | Record the gap and keep completion open |

Render a source-backed WideField Security adoption packet and delegate target
work to the child skill that owns each system.

## Workflow

1. Read `reference.md` for source boundaries and unsupported API rules.
2. Copy `template.example` to a local-only spec if the user has many non-secret
   values to track.
3. Render the parent packet:

```bash
bash skills/widefield-security-setup/scripts/setup.sh --render
```

4. Delegate child render/validate execution with the router only after review:

```bash
bash skills/widefield-security-setup/scripts/setup.sh --apply --accept-apply \
  --children okta,splunk,doctor
```

Saviynt and Google SecOps child validation is evidence-gated. To include either
child, pass `--evidence-file` containing its required marker. A combined JSON
file may be used when routing both children, but it must contain both a Saviynt
remediation outcome marker and Google SecOps log type `WIDEFIELD_SECURITY`:

```bash
bash skills/widefield-security-setup/scripts/setup.sh --apply --accept-apply \
  --children saviynt,google \
  --evidence-file ./widefield-cross-system-evidence.local.json
```

The parent never calls private or undocumented WideField APIs. Parent apply is
limited to child render/validate orchestration. Live mutation is limited to
child skills that explicitly document supported public API paths and are run
directly with their required file-backed credentials.

## Guardrails

- Keep secrets in files; reject raw token, password, API key, and client secret
  arguments.
- Treat WideField platform configuration as a provider/customer handoff unless
  public API coverage is added to `reference.md`.
- Run `validate.sh --dry-run` before target-specific validation.
- Parent delegation forwards optional target values only when they were
  explicitly supplied; it never synthesizes empty child arguments.

Files in this skill

  • SKILL.md5.1 KB
  • agents/openai.yaml262 B
  • reference.md3.1 KB
  • scripts/render_assets.py2.2 KB
  • scripts/setup.sh422 B
  • scripts/validate.sh424 B
  • template.example810 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.