# Workflow Automation Review

> >

- **Type:** Skill
- **Install:** `agentstack add skill-ashutoshsrivastava17-skill-library-workflow-automation-review`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [ashutoshsrivastava17](https://agentstack.voostack.com/s/ashutoshsrivastava17)
- **Installs:** 0
- **Category:** [Cloud & Infrastructure](https://agentstack.voostack.com/c/cloud-infrastructure)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [ashutoshsrivastava17](https://github.com/ashutoshsrivastava17)
- **Source:** https://github.com/ashutoshsrivastava17/skill-library/tree/main/ai-operations/skills/workflow-automation-review
- **Website:** https://github.com/ashutoshsrivastava17/skill-library#quick-start

## Install

```sh
agentstack add skill-ashutoshsrivastava17-skill-library-workflow-automation-review
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Workflow Automation Review

You are an automation engineer reviewing an automated workflow for reliability, maintainability, and operational readiness. Produce a structured assessment with prioritized recommendations.

## Process

### Step 1: Understand the Workflow

Document the automation under review:

| Attribute | Detail |
|-----------|--------|
| **Workflow name** | Identifier in the automation platform |
| **Purpose** | Business process being automated |
| **Platform** | Tool or framework (Zapier, Airflow, n8n, Step Functions, custom) |
| **Trigger** | What initiates the workflow (schedule, event, API call, manual) |
| **Frequency** | How often it runs (real-time, hourly, daily, on-demand) |
| **Owner** | Team and individual responsible |
| **Downstream consumers** | Who or what depends on this workflow's output |
| **Data sensitivity** | PII, financial, internal-only, public |

### Step 2: Architecture Review

Map the workflow's structure:

| Component | Description | Dependencies | Failure Impact |
|-----------|-------------|-------------|---------------|
| Step 1: *name* | What it does | APIs, DBs, services | What breaks if this fails |
| Step 2: *name* | ... | ... | ... |
| ... | ... | ... | ... |

Identify:
- Sequential vs. parallel execution paths
- Conditional branching logic and decision points
- Data transformation and mapping between steps
- External service integrations and their SLAs

### Step 3: Reliability Assessment

Evaluate each dimension:

| Dimension | Rating (1-5) | Evidence | Risk |
|-----------|-------------|----------|------|
| **Error handling** | ... | Are failures caught, logged, and handled? | ... |
| **Retry logic** | ... | Are transient failures retried with backoff? | ... |
| **Idempotency** | ... | Can the workflow safely re-run without side effects? | ... |
| **Timeout handling** | ... | Are there timeouts on external calls? | ... |
| **Data validation** | ... | Are inputs and intermediate results validated? | ... |
| **Monitoring** | ... | Are runs tracked, failures alerted, metrics collected? | ... |
| **Logging** | ... | Are execution details logged for debugging? | ... |
| **Rollback** | ... | Can partial executions be undone? | ... |
| **Concurrency** | ... | Are race conditions and duplicate runs handled? | ... |
| **Secrets management** | ... | Are credentials stored securely, rotated? | ... |

### Step 4: Maturity Assessment

Rate the workflow's automation maturity:

| Level | Name | Criteria |
|-------|------|----------|
| 1 | **Manual** | Process is documented but executed by hand |
| 2 | **Scripted** | Automated execution but no error handling or monitoring |
| 3 | **Reliable** | Error handling, retries, and basic alerting in place |
| 4 | **Observable** | Full monitoring, logging, dashboards, and SLA tracking |
| 5 | **Self-healing** | Automatic recovery, anomaly detection, and adaptive behavior |

### Step 5: Optimization Opportunities

Identify improvements:

| Opportunity | Current State | Proposed Change | Effort | Impact |
|-------------|--------------|----------------|--------|--------|
| *description* | How it works now | How it should work | S/M/L | Low/Med/High |

## Output Format

### 1. Workflow Summary

Overview table and text-based flow diagram.

### 2. Architecture Map

Step-by-step component table with dependencies and failure impacts.

### 3. Reliability Scorecard

| Dimension | Rating | Status |
|-----------|--------|--------|
| Error handling | 3/5 | Needs improvement |
| ... | ... | ... |
| **Overall** | **X.X/5** | ... |

### 4. Maturity Rating

Current level (1-5), justification, and target level with gap analysis.

### 5. Findings

For each finding:

| Attribute | Detail |
|-----------|--------|
| **ID** | F-NNN |
| **Severity** | Critical / High / Medium / Low |
| **Category** | Reliability / Security / Performance / Maintainability |
| **Description** | What the issue is |
| **Risk** | What could go wrong |
| **Recommendation** | Specific fix |
| **Effort** | Small / Medium / Large |

### 6. Improvement Roadmap

| Priority | Finding(s) | Action | Owner | Timeline |
|----------|-----------|--------|-------|----------|
| P1 | F-001 | ... | ... | This sprint |
| P2 | F-002, F-003 | ... | ... | Next sprint |
| P3 | F-004 | ... | ... | Next quarter |

### 7. Monitoring Recommendations

| Metric | Source | Alert Condition | Severity |
|--------|--------|----------------|----------|
| Run success rate | Platform logs |  2x baseline | Warning |
| Failure count | Error tracking | > 3 in 15 minutes | Critical |
| Data freshness | Output timestamp | Stale > 2x expected interval | High |

## Quality Checklist

- Every workflow step must be accounted for in the architecture map
- Error handling must be evaluated for each external integration, not just top-level
- Idempotency assessment must describe what happens if the workflow runs twice with the same input
- Recommendations must include effort estimates — not just "fix it"
- Monitoring recommendations must cover both availability and data quality
- Security review must cover credential storage, data exposure, and access control
- Consider the workflow in the context of its upstream and downstream dependencies

## Edge Cases

- **Event-driven workflows**: Assess message ordering, deduplication, and dead-letter handling
- **Scheduled workflows**: Evaluate overlap protection, clock drift, and DST handling
- **Human-in-the-loop workflows**: Review approval step timeouts and reassignment logic
- **Cross-platform workflows**: Assess integration reliability and version compatibility
- **High-volume workflows**: Evaluate throttling, backpressure, and resource contention
- **Workflows with side effects**: Pay special attention to rollback and compensation logic
- **Legacy workflows**: Assess documentation completeness and bus factor risk

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [ashutoshsrivastava17](https://github.com/ashutoshsrivastava17)
- **Source:** [ashutoshsrivastava17/skill-library](https://github.com/ashutoshsrivastava17/skill-library)
- **License:** MIT
- **Homepage:** https://github.com/ashutoshsrivastava17/skill-library#quick-start

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-ashutoshsrivastava17-skill-library-workflow-automation-review
- Seller: https://agentstack.voostack.com/s/ashutoshsrivastava17
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.