Works with the AI Agents you already use
data product opportunity assessor
Data initiatives often stall because of unverified value claims, missing data rights, or technical debt hidden in the data lineage.
$2.99
data product opportunity assessor
Example session with this skill installed
Assess the opportunity for a 'Customer Churn Predictor' model using 'crm_export_v1' and 'web_logs_2023'. We claim a $200k annual saving based on a stakeholder estimate. No experiment data yet.
- Read your context and instructions
- Compiled the data product opportunity
I have registered the Customer Churn Predictor use case and indexed the provided CRM and web log assets. I ran the assessment script to score the opportunity based on the stakeholder estimate provided.
Data Product Opportunity Assessment — Customer Churn Predictor, 2024-05-20
1. Scope, sources and assumptions
- Sources: E1 (User Prompt: Stakeholder assertion on savings).
- Assumptions: Default weights applied; threshold for 'Ready' set at 70/100.
2. Opportunity scorecard
| Dimension | Score | Weight |
|---|---|---|
| Value | 15.0 | 0.40 |
| Feasibility | 45.0 | 0.20 |
| Readiness | 30.0 | 0.20 |
| Risk | 55.0 | 0.20 |
| Composite | 31.0 | 1.00 |
Recommendation: HOLD. Missing evidence for value claims and unverified data rights are blocking the gate.
3. Value claims and evidence
| Claim | Annual value | Class | Evidence |
|---|---|---|---|
| Churn Reduction | $200,000 | Estimated | E1 (Stakeholder Assertion) |
- Validated Total: $0
- Estimated Total: $200,000
- NPV: Undefined (Insufficient validated value)
4. Data readiness, rights and permissions
| Asset | Eligible | Reasons | Readiness contribution |
|---|---|---|---|
| crm_export_v1 | No | rights.status unknown | 0% |
| web_logs_2023 | No | access_granted false | 0% |
5. Feasibility and risk
- Feasibility: Low. Required data assets are currently marked as ineligible.
- Risk Register:
- R1: Data Access (Likelihood 5, Impact 4)
- R2: Value Realization (Likelihood 4, Impact 5)
6. Dependency map
crm_export_v1->Churn Predictor(Input)web_logs_2023->Churn Predictor(Input)Churn Reduction Claim->Churn Predictor(Outcome)
7. Pilot charter (draft)
- Objective: Predict high-risk churn customers to trigger retention offers.
- In-scope assets:
crm_export_v1,web_logs_2023. - Success Metrics: Model Precision > 0.7, 10% reduction in churn in test group.
- Entry Conditions: Requires
access_grantedfor all inputs and a privacy approval reference.
8. Gaps, warnings and items for human review
- Information gap — organizational validation required: Rights status for CRM data is missing.
- Analytical estimate — validation required: The $200k value claim lacks historical analogue or measured evidence.
- Warning: Composite score is below the 50% threshold for pilot commencement.
This assessment supports, and does not replace, management's decision.
Next steps
- Provide the privacy approval reference and access confirmation for the CRM and web log datasets.
- Link the value claim to a historical analogue or benchmark to move it from 'Estimated' to 'Validated'.
- Define the specific features in
crm_export_v1to check for target leakage before build.
Connects securely to your tools. The creator never sees your data.
About this skill
Data initiatives often stall because of unverified value claims, missing data rights, or technical debt hidden in the data lineage. This skill subjects proposed datasets and models to a deterministic audit to verify if an analytics opportunity is actually viable before you commit budget or engineering hours.
What it does
- Evidence-backed scoring calculates value, feasibility, and readiness scores using a deterministic script.
- Risk and leakage auditing flags feature leakage, training-test overlap, and data assets lacking explicit approval or access rights.
- Value validation separates validated causal claims from mere estimates and unsupported assertions.
- Dependency mapping visualizes the relationships between data assets, stakeholders, and value claims to identify bottlenecks.
- Pilot chartering generates a structured draft charter including success metrics, budget requirements, and go/no-go entry conditions.
How it works
- Supply context by providing a use case description, purpose, and a list of available data assets.
- Register evidence as the agent indexes your documents, benchmarks, and stakeholder assertions to back value claims.
- Run deterministic audit via a provided Python script that processes the input schema to ensure no hallucinated scores.
- Review the assessment which includes a scorecard, dependency map, and a risk register.
Frameworks & tools
The skill utilizes a custom Python-based assessment engine (run.py) and JSON-schema validation. It operates independently of specific data stacks, requiring only structured metadata about your assets.
Why this beats prompting it yourself
Generic prompts often hallucinate "common sense" benchmarks or ignore the lack of legal rights to a dataset. This skill uses a strict logic gate that withholds credit for any claim lacking evidence and automatically excludes assets without verified access or privacy approval references.
Use cases
- Pilot prioritization to decide which data products in a backlog deserve funding based on readiness and risk.
- Due diligence for checking if a proposed model will suffer from feature leakage or missing data rights before development starts.
- Stakeholder alignment by providing an objective scorecard that maps specific business needs to data asset availability.
Known limitations
Does not perform raw data quality monitoring or privacy re-identification analysis. Requires manual input of asset metadata as it does not crawl your internal databases directly.
Requires metadata about data assets (approvals, rights, sensitivity). Does not perform direct PII scanning or raw data profiling. Decision support only; cannot approve budgets or legal use.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 days ago