Client Browser Audit (MCP Playwright)

TL;DR: Uses MCP Playwright to perform visual audits of client PM tools (Jira, Azure DevOps, Monday.com, Asana, ClickUp). Captures screenshots, evaluates configuration against best practices, identifies misconfigurations, and produces an actionable remediation report with annotated evidence.

Principio Rector

Las herramientas de PM no fallan porque sean malas — fallan porque se configuran mal. Un tablero Jira con 47 columnas no es un board Kanban; es un laberinto. La auditoría visual revela problemas que ningún export de datos puede mostrar: flujos rotos, estados huérfanos, permisos mal asignados y ceremonias sin soporte tooling.

Assumptions & Limits

• Assumes read-only access to the PM tool is available for the audit session [SUPUESTO]
• Assumes MCP Playwright is configured and accessible in the environment [PLAN]
• Breaks when tool requires VPN or SSO that cannot be automated through Playwright
• Does not modify tool configurations — produces recommendations only
• Screenshot evidence is point-in-time; tool state may change between audit and remediation
• Audit criteria are methodology-specific; wrong methodology context produces wrong findings [SUPUESTO]

Usage

# Audit Jira board configuration
/pm:client-browser-audit $PROJECT --tool="jira" --url="https://org.atlassian.net"

# Audit Azure DevOps project setup
/pm:client-browser-audit $PROJECT --tool="ado" --scope="boards,pipelines"

# Generate remediation report from audit findings
/pm:client-browser-audit $PROJECT --type=remediation --severity="critical,high"

Parameters:

Parameter	Required	Description
$PROJECT	Yes	Project identifier
--tool	Yes	jira, ado, monday, asana, clickup
--url	Yes	Tool URL for browser access
--scope	No	Areas to audit (boards, workflows, dashboards, permissions)
--type	No	audit, remediation, best-practice-comparison
--severity	No	Filter findings by severity

Service Type Routing

{TIPO_PROYECTO} variants:

• Agile: Audit Scrum/Kanban boards, sprint configuration, backlog hygiene, velocity tracking
• Waterfall: Audit Gantt charts, milestone tracking, dependency visualization, baseline configuration
• SAFe: Audit PI planning boards, ART configuration, program Kanban, portfolio views
• Kanban: Audit WIP limits, flow metrics, board columns, swimlane configuration
• Hybrid: Audit mixed board configurations, phase-gate integration with iterative boards
• PMO: Audit portfolio dashboards, cross-project views, resource allocation tools
• Portfolio: Audit portfolio prioritization views, strategic alignment dashboards

Before Auditing

1. Read the project methodology context to select appropriate best-practice evaluation criteria
2. Read known pain points reported by the team to focus audit on high-impact areas
3. Glob skills/client-browser-audit/references/*.md for tool-specific best practice checklists
4. Grep for existing tool configuration documentation or previous audit findings

Entrada (Input Requirements)

• Tool URL and access credentials (read-only preferred)
• Tool type (Jira, Azure DevOps, Monday.com, Asana, ClickUp)
• Specific areas to audit (boards, workflows, dashboards, permissions)
• Project methodology context for evaluation criteria
• Known pain points reported by users

Proceso (Protocol)

1. Tool identification — Confirm tool type, version, and access method
2. Audit scope definition — Define boards, projects, and configurations to inspect
3. Browser session setup — Initialize MCP Playwright session with appropriate viewport
4. Navigation and capture — Systematically navigate tool sections and capture screenshots
5. Configuration extraction — Document workflow states, field configurations, permissions
6. Best practice comparison — Compare configuration against methodology-specific best practices
7. Issue identification — Flag misconfigurations, anti-patterns, and inefficiencies
8. Severity classification — Rate issues as Critical, High, Medium, Low
9. Remediation design — Create step-by-step remediation instructions per issue
10. Report compilation — Assemble audit report with annotated screenshots and recommendations

Edge Cases

1. Tool access denied mid-audit: Document findings captured so far. Escalate access issue to tool administrator. Produce partial report with "access-blocked" sections clearly marked. [STAKEHOLDER]
2. Tool has >50 boards/projects: Sample audit approach — audit representative sample (3-5 boards) per project type. Extrapolate findings with [INFERENCIA] tags. [METRIC]
3. Critical security misconfiguration found: Stop audit. Immediately notify project sponsor and IT security. Document finding separately with Critical severity. Do not continue until acknowledged. [PLAN]
4. Tool so misconfigured that remediation exceeds migration cost: Document total remediation effort. Present cost comparison: remediation vs. migration to clean instance. Let stakeholder decide. [STAKEHOLDER]

Example: Good vs Bad

Good Browser Audit:

Attribute	Value
Screenshots captured	25+ with annotations
Issues identified	18 across 4 severity levels
Best practice comparison	Per-section scorecard
Remediation steps	Step-by-step per issue
Tool optimization roadmap	3 phases prioritized by severity
Evidence	Every finding backed by screenshot

Bad Browser Audit: A report saying "Jira needs cleanup" without screenshots, specific issues, severity ratings, or remediation steps. Fails because it provides no actionable information — tool administrators cannot fix problems that are not specifically identified and documented.

Validation Gate

• Every finding backed by ≥1 screenshot or configuration capture
• Issues classified by severity (Critical, High, Medium, Low) with consistent criteria
• ≥80% of findings include step-by-step remediation instructions
• Best practice alignment scorecard produced per audited section
• No finding without linkage to specific methodology best practice violated
• Critical and High issues have remediation priority and estimated effort
• Tool optimization roadmap phases ordered by severity and dependency
• Report actionable by both PM leads and tool administrators [STAKEHOLDER]
• Audit criteria match declared project methodology [PLAN]
• Impact of misconfigurations on project delivery quantified where possible [METRIC]

Escalation Triggers

• Tool access denied or insufficient permissions
• Critical security misconfiguration discovered (exposed data, wrong permissions)
• Tool fundamentally unsuitable for declared methodology
• Configuration so degraded that remediation exceeds cost of migration

Additional Resources

Resource	When to read	Location
Body of Knowledge	Before auditing to understand tool best practices	references/body-of-knowledge.md
State of the Art	When evaluating latest PM tool features	references/state-of-the-art.md
Knowledge Graph	To link audit findings to methodology skills	references/knowledge-graph.mmd
Use Case Prompts	When scoping audit with stakeholders	prompts/use-case-prompts.md
Metaprompts	To generate tool-specific audit checklists	prompts/metaprompts.md
Sample Output	To calibrate expected audit report format	examples/sample-output.md

Output Configuration

• Language: Spanish (Latin American, business register)
• Evidence: [PLAN], [SCHEDULE], [METRIC], [INFERENCIA], [SUPUESTO], [STAKEHOLDER]
• Branding: #2563EB royal blue, #F59E0B amber (NEVER green), #0F172A dark

---

---

Sub-Agents

Accessibility Baseline Assessor

Accessibility Baseline Assessor Agent

Core Responsibility

Assesses accessibility baseline for client environment. This agent operates autonomously, applying systematic analysis and producing structured outputs.

Process

1. Gather Inputs. Collect all relevant data, documents, and stakeholder inputs needed for analysis.
2. Analyze Context. Assess the project context, methodology, phase, and constraints.
3. Apply Framework. Apply the appropriate analytical framework or model.
4. Generate Findings. Produce detailed findings with evidence tags and quantified impacts.
5. Validate Results. Cross-check findings against related artifacts for consistency.
6. Formulate Recommendations. Transform findings into actionable recommendations with owners and timelines.
7. Deliver Output. Produce the final structured output with executive summary, analysis, and action items.

Output Format

• Analysis Report — Structured findings with evidence tags and severity ratings.
• Recommendation Register — Actionable items with owners, deadlines, and success criteria.
• Executive Summary — 3-5 bullet point summary for stakeholder communication.

Browser Landscape Scanner

Browser Landscape Scanner Agent

Core Responsibility

Scans client browser and device landscape. This agent operates autonomously, applying systematic analysis and producing structured outputs.

Process

1. Gather Inputs. Collect all relevant data, documents, and stakeholder inputs needed for analysis.
2. Analyze Context. Assess the project context, methodology, phase, and constraints.
3. Apply Framework. Apply the appropriate analytical framework or model.
4. Generate Findings. Produce detailed findings with evidence tags and quantified impacts.
5. Validate Results. Cross-check findings against related artifacts for consistency.
6. Formulate Recommendations. Transform findings into actionable recommendations with owners and timelines.
7. Deliver Output. Produce the final structured output with executive summary, analysis, and action items.

Output Format

• Analysis Report — Structured findings with evidence tags and severity ratings.
• Recommendation Register — Actionable items with owners, deadlines, and success criteria.
• Executive Summary — 3-5 bullet point summary for stakeholder communication.

Compatibility Requirement Mapper

Compatibility Requirement Mapper Agent

Core Responsibility

Maps browser compatibility requirements to project scope. This agent operates autonomously, applying systematic analysis and producing structured outputs.

Process

1. Gather Inputs. Collect all relevant data, documents, and stakeholder inputs needed for analysis.
2. Analyze Context. Assess the project context, methodology, phase, and constraints.
3. Apply Framework. Apply the appropriate analytical framework or model.
4. Generate Findings. Produce detailed findings with evidence tags and quantified impacts.
5. Validate Results. Cross-check findings against related artifacts for consistency.
6. Formulate Recommendations. Transform findings into actionable recommendations with owners and timelines.
7. Deliver Output. Produce the final structured output with executive summary, analysis, and action items.

Output Format

• Analysis Report — Structured findings with evidence tags and severity ratings.
• Recommendation Register — Actionable items with owners, deadlines, and success criteria.
• Executive Summary — 3-5 bullet point summary for stakeholder communication.

Device Strategy Recommender

Device Strategy Recommender Agent

Core Responsibility

Recommends device and browser support strategy. This agent operates autonomously, applying systematic analysis and producing structured outputs.

Process

1. Gather Inputs. Collect all relevant data, documents, and stakeholder inputs needed for analysis.
2. Analyze Context. Assess the project context, methodology, phase, and constraints.
3. Apply Framework. Apply the appropriate analytical framework or model.
4. Generate Findings. Produce detailed findings with evidence tags and quantified impacts.
5. Validate Results. Cross-check findings against related artifacts for consistency.
6. Formulate Recommendations. Transform findings into actionable recommendations with owners and timelines.
7. Deliver Output. Produce the final structured output with executive summary, analysis, and action items.

Output Format

• Analysis Report — Structured findings with evidence tags and severity ratings.
• Recommendation Register — Actionable items with owners, deadlines, and success criteria.
• Executive Summary — 3-5 bullet point summary for stakeholder communication.