Governance Readiness Assessor

Prerequisites

Required MCP Servers: All 6 AAP MCP servers

• aap-mcp-job-management
• aap-mcp-inventory-management
• aap-mcp-configuration
• aap-mcp-security-compliance
• aap-mcp-system-monitoring
• aap-mcp-user-management

Verification: Run the aap-mcp-validator skill with all 6 servers before proceeding.

When to Use This Skill

Use this skill when:

• Full assessment: User asks to assess or audit AAP governance readiness, or asks if AAP is ready for production
• Scoped assessment: User asks about a specific governance area:
  • "Assess my credentials setup" → Domain 4 (Credential Security)
  • "Check my RBAC" / "How are my users and teams configured?" → Domain 3 (Access Control)
  • "Are my notifications configured properly?" → Domain 2 (Notification Coverage)
  • "Do I have workflows set up?" → Domain 1 (Workflow Governance)
  • "Check my execution environments" → Domain 5 (Execution Environments)
  • "Assess workload isolation" → Domain 6 (Workload Isolation)
  • "Is my audit trail working?" → Domain 7 (Audit Trail)
  • "Check my authentication setup" → Bonus (External Authentication)
  • Combinations: "Assess my credentials and RBAC" → Domains 3 + 4
• Before a first production execution (as part of governance-assessor workflow)
• User asks what needs to be fixed in their AAP configuration

Do NOT use when:

• User wants to execute a specific job (use execution-risk-analyzer + governed-job-launcher skills)
• User wants to troubleshoot a failed job (use job-failure-analyzer skill)
• User only wants MCP connectivity validation (use aap-mcp-validator skill)

Workflow

Step 1: Consult Governance Readiness Documentation

CRITICAL: Document consultation MUST happen BEFORE any MCP tool invocations.

Document Consultation (REQUIRED - Execute FIRST):

1. Action: Read governance-readiness.md using the Read tool to understand the 7-domain assessment framework, Red Hat source citations, decision tables, and output template
2. Output to user: "I consulted governance-readiness.md to understand Red Hat's governance best practices for the 7-domain assessment framework."

Step 1.5: Determine Assessment Scope

Analyze the user's request to determine whether this is a full assessment (all 7+1 domains) or a scoped assessment (specific domains).

Scope Detection:

Rules:

• If scope is ambiguous, default to full assessment
• For scoped assessments, still validate only the MCP servers needed for the requested domains
• For scoped assessments, still perform cross-domain correlation IF the assessed domains participate in any correlation pattern (e.g., assessing credentials alone still triggers the RBAC+Credentials correlation check by querying enough RBAC data to determine if teams exist)
• Always tell the user what scope was detected: "You asked about credentials -- I'll assess Domain 4 (Credential Security) and check for related cross-domain impacts."

Step 2: Query MCP Servers (Scoped or Full)

For full assessment: Execute all queries below in parallel where possible. For scoped assessment: Execute only the queries for the requested domains. However, if the requested domains participate in a cross-domain correlation pattern (see Step 5), also execute the minimal queries needed for that correlation (e.g., assessing credentials alone should also query teams_list to check if the RBAC+Credentials correlation applies).

Domain 1 - Workflow Governance:

MCP Tool: workflow_job_templates_list (from aap-mcp-job-management) Parameters:

• page_size: 100

MCP Tool: job_templates_list (from aap-mcp-job-management) Parameters:

• page_size: 100

Domain 2 - Notification Coverage:

MCP Tool: notification_templates_list (from aap-mcp-configuration) Parameters:

• page_size: 100

Domain 3 - Access Control (RBAC):

MCP Tool: users_list (from aap-mcp-user-management) Parameters:

• page_size: 100

MCP Tool: teams_list (from aap-mcp-user-management) Parameters:

• page_size: 100

MCP Tool: role_user_assignments_list (from aap-mcp-user-management) Parameters:

• page_size: 100

MCP Tool: role_team_assignments_list (from aap-mcp-user-management) Parameters:

• page_size: 100

Domain 4 - Credential Security:

MCP Tool: credentials_list (from aap-mcp-security-compliance) Parameters:

• page_size: 100

MCP Tool: credential_types_list (from aap-mcp-security-compliance) Parameters:

• page_size: 100

Domain 5 - Execution Environments:

MCP Tool: execution_environments_list (from aap-mcp-configuration) Parameters:

• page_size: 100

Domain 6 - Workload Isolation:

MCP Tool: instance_groups_list (from aap-mcp-system-monitoring) Parameters:

• page_size: 100

Domain 7 - Audit Trail:

MCP Tool: activity_stream_list (from aap-mcp-system-monitoring) Parameters:

• page_size: 10

Bonus - External Authentication:

MCP Tool: authenticators_list (from aap-mcp-user-management) Parameters:

• page_size: 100

Step 3: Assess Each Domain (with Mandatory Red Hat Citations)

For EACH assessed domain (all 7+1 for full assessment, or the scoped domains), follow the assessment pattern from governance-readiness.md.

MANDATORY CITATION RULE: Every single domain finding MUST include:

1. The Red Hat source document name (e.g., "Configuring Automation Execution")
2. The specific chapter and section (e.g., "Ch. 15, Sec. 15.1.4")
3. A direct quote from the Red Hat documentation
4. The recommendation must reference the Red Hat source again

A finding without a Red Hat citation is incomplete. Every assessment criterion and recommendation must trace to official Red Hat documentation.

Output format per domain (MANDATORY -- do not deviate):

### Domain [N]: [Name] — [PASS/GAP/WARN]

Per Red Hat's *[Source Document Title]* ([Chapter/Section]):
> "[Direct quote from Red Hat documentation]"

**Finding**: [What the MCP query revealed -- specific numbers, names, counts]
**Status**: [PASS / GAP / WARN] -- [one-line rationale from decision table]
**Recommendation**: Per Red Hat's *[Source]* ([Chapter/Section]): [specific action to take]
**Source URL**: [full URL to the Red Hat documentation page]

Citation verification checklist (self-check before outputting each domain):

• Does this finding include a direct Red Hat quote?
• Does the recommendation cite a specific Red Hat source (not just "Red Hat recommends")?
• Is the chapter/section number included (not just the document title)?
• Would someone reading this output know exactly which Red Hat doc page to visit?

For scoped assessments: Apply the same citation rigor. Even if assessing a single domain, the output must include the full Red Hat source attribution.

Step 4: Adaptive Depth Queries (Scenario-Driven)

After the initial per-domain assessment, perform follow-up queries based on what was discovered. These are NOT executed for every assessment -- they fire only when specific conditions are met, adapting the assessment to the scenario.

Per the "Adaptive Depth Queries" section of governance-readiness.md:

4a. If Domain 2 (Notifications) is PASS: Verify notification templates are actually bound to job templates, not just existing unused. Query job_templates_list and check notification association fields. If all templates have empty notification bindings, downgrade Domain 2 to WARN.

4b. If Domain 4 (Credentials) is PASS (multiple credentials): Check if the same credential is shared across templates targeting different inventories. Query job_templates_list and compare credential fields. If one credential spans dev and prod templates, downgrade Domain 4 to WARN with a separation-of-duties finding.

4c. If Domain 3 (RBAC) is PASS (teams exist): Check whether team role assignments follow least privilege. Query role_team_assignments_list in detail. If any team has Admin-level access on organization-wide scope, downgrade Domain 3 to WARN.

4d. Scale Calibration (always): Query inventories_list and hosts_list to determine environment scale. Calibrate severity framing: small lab (< 5 hosts, dev-only) vs enterprise (> 50 hosts, production inventories). Include the calibration context in the report preamble.

Step 5: Cross-Domain Correlation

After all domains are assessed (including depth adjustments from Step 4), check for compound risks per the "Cross-Domain Correlation" section of governance-readiness.md:

5a. RBAC GAP + Credentials exist: Flag that without teams, credentials are necessarily user-scoped. Recommend fixing RBAC first.

5b. No Workflows GAP + No Notifications GAP: Flag the highest-risk combination -- no governance controls AND no visibility on production failures.

5c. Single Instance Group + production and dev inventories: Flag shared execution capacity between production and development workloads.

5d. Multiple superusers + local-only auth: Flag that superuser accounts without MFA/external auth have maximum blast radius.

Include compound findings in the report as a dedicated section AFTER the domain-by-domain findings.

Step 6: Generate Summary Report

For full assessment: Produce the complete Governance Readiness Report following the output template in governance-readiness.md. Include:

1. Assessment date and AAP instance identifier
2. Scale calibration context (lab vs enterprise)
3. Per-domain findings with Red Hat citations (reflecting any depth-query adjustments from Step 4)
4. Compound Risk Analysis section (from Step 5 cross-correlation)
5. Summary table with all domain statuses
6. Overall score: X PASS, Y WARN, Z GAP out of 8 domains
7. Recommended Fix Order -- prioritized by dependency chain per governance-readiness.md
8. Sources consulted -- list all Red Hat documentation URLs referenced in the report

For scoped assessment: Produce a focused report for the requested domains:

1. Assessment date, scope statement ("Assessing: Credential Security"), and AAP instance
2. Scale calibration context
3. Per-domain findings for the scoped domains only (same mandatory citation format)
4. Depth-query results if triggered
5. Cross-domain correlations that involve the assessed domains (if any)
6. Status summary for assessed domains
7. Sources consulted -- list Red Hat documentation URLs referenced
8. Offer: "Would you like me to run the full 7-domain assessment for complete coverage?"

Step 7: Offer Remediation

For any domains with GAP or WARN status, offer to remediate using MCP write tools where available:

Human-in-the-Loop: Before creating or modifying any resource, present the planned change and ask for explicit user approval.

Dependencies

Required MCP Servers

• aap-mcp-job-management - Workflow and job template data
• aap-mcp-inventory-management - Inventory data
• aap-mcp-configuration - Notifications, EEs, settings
• aap-mcp-security-compliance - Credentials and credential types
• aap-mcp-system-monitoring - Instance groups and activity stream
• aap-mcp-user-management - Users, teams, roles, authenticators

Required MCP Tools

• workflow_job_templates_list, job_templates_list (from job-management)
• notification_templates_list (from configuration)
• execution_environments_list (from configuration)
• users_list, teams_list, role_user_assignments_list, role_team_assignments_list, authenticators_list (from user-management)
• credentials_list, credential_types_list (from security-compliance)
• instance_groups_list, activity_stream_list (from system-monitoring)

Related Skills

• aap-mcp-validator - Prerequisite: validate all 6 servers first
• execution-summary - Generate audit trail after assessment

Reference Documentation

• governance-readiness.md - The 7-domain assessment reference

Critical: Human-in-the-Loop Requirements

This skill requires explicit user confirmation at the following steps:

1. Before Remediation Actions

   • Display the planned change (e.g., "Create team 'automation-operators'")
   • Ask: "Should I create this resource to address the gap?"
   • Wait for explicit user confirmation

2. Never auto-remediate: Always present findings first and wait for user decision on remediation

Example Usage

Example 1: Full Assessment

User: "Assess my AAP platform's governance readiness"

Agent:

1. Reads governance-readiness.md
2. Scope detection: full assessment (all 7+1 domains)
3. Validates all 6 MCP servers via aap-mcp-validator
4. Queries each server (Step 2)
5. Assesses each domain -- every finding includes Red Hat quote, chapter/section, and source URL
6. Adapts: Notifications exist but aren't bound to any templates → downgrades Domain 2 from PASS to WARN
7. Adapts: Only 3 hosts in a single dev inventory → calibrates severity
8. Correlates: No workflows AND no notifications → compound risk flagged
9. Produces full report with Red Hat sources list at the bottom
10. Offers remediation, starting with highest-priority gap

Example 2: Scoped Assessment

User: "Can you assess my credentials setup in the AAP?"

Agent:

1. Reads governance-readiness.md (specifically Domain 4: Credential Security)

2. Scope detection: Domain 4 (Credential Security). Also queries minimal RBAC data for cross-correlation.

3. Reports: "You asked about credentials -- I'll assess Domain 4 (Credential Security) and check for related cross-domain impacts."

4. Validates aap-mcp-security-compliance and aap-mcp-user-management (for correlation)

5. Queries credentials_list, credential_types_list, and teams_list (for correlation)

6. Assesses with Red Hat citations:

   ### Domain 4: Credential Security — WARN
   
   Per Red Hat's *Configuring Automation Execution* (Ch. 15, Sec. 15.1.5):
   > "Enforce separation of duties...different credentials for different pieces of automation."
   
   **Finding**: Found 2 credentials, both of type "Machine." Credential "ssh-key" (ID: 3) is used across both dev and production job templates.
   **Status**: WARN -- credentials exist but lack separation of duties
   **Recommendation**: Per Red Hat's *Security Best Practices* (Ch. 15, Sec. 15.1.5): Create separate credentials per environment.
   **Source URL**: https://docs.redhat.com/en/.../configuring_automation_execution/controller-security-best-practices

7. Correlates: Checks RBAC -- no teams exist → flags compound: "Without teams, credentials are necessarily user-scoped (Ch. 15, Sec. 15.1.4). Fix RBAC first to enable team-scoped credential management."

8. Offers: "Would you like me to run the full 7-domain assessment for complete coverage?"