Works with the AI Agents you already use
data source discovery and profiling
Data sources are often undocumented, stale, or lack clear ownership, leading to wasted effort and flawed analysis.
$2.99
data source discovery and profiling
Example session with this skill installed
Profile the provided inventory_v1.csv and user_api_meta.json. Check for completeness, freshness against an as-of date of 2024-05-20, and tell me if they are fit for 'Quarterly Revenue Analysis'.
- Read your context and instructions
- Compiled the data source discovery
- Generated the data export
I have processed the inventory metadata and sample records for the two provided sources. The profiling script successfully calculated completeness, validity, and freshness metrics based on the 2024-05-20 as-of date.
Data Source Discovery and Profiling — Finance Dept, 2024-05-20
1. Scope, sources and assumptions
Sources processed: inventory_v1.csv (E1), user_api_meta.json (E2). Analysis purpose: Quarterly Revenue Analysis. Assumptions: Sample records are representative of the source population.
2. Source inventory
| Source | Kind | Owner | Access | Scope | Approved | Last updated | Freshness |
|---|---|---|---|---|---|---|---|
| REV_24_Q1 | file | Sarah Chen | granted | [*] | yes | 2024-05-19 | current |
| USER_API | api | n/a | pending | [id, email] | no | 2024-05-10 | stale |
3. Profiling scorecard
| Source | Fitness index | Completeness | Validity | Uniqueness | Freshness | Documentation |
|---|---|---|---|---|---|---|
| REV_24_Q1 | 0.9500 | 0.9800 | 0.9400 | 1.0000 | 1.0000 | 0.8500 |
| USER_API | 0.4200 | 0.8000 | 0.6000 | 0.9000 | 0.3000 | 0.5000 |
Note: Indices are fractions (0.9500 means 95 percent). REV_24_Q1 shows 2% missing values in transaction_id.
4. Suitability report
| Source | Verdict | Blockers | Conditions |
|---|---|---|---|
| REV_24_Q1 | suitable | none | none |
| USER_API | unsuitable | stale_data, pending_access | requires owner approval |
5. Access, scope and ownership findings
USER_API is currently
Information gap — organizational validation required regarding ownership. Access for USER_API is pending.
6. Warnings and limitations
- W042: Sample size for USER_API is below
min_sample_rows. - Profiles describe samples only; they do not guarantee future data quality.
7. Items for human review and proposed actions
- Management decision required: Approve access for USER_API.
- Analytical estimate — validation required: REV_24_Q1 is fit for revenue analysis based on 98% completeness.
- Proposal: Archive USER_API if freshness cannot be improved.
This analysis supports, and does not replace, professional judgment and management's decision.
Next steps
- Review the suitability blockers for USER_API with the IT team.
- Verify the
transaction_iduniqueness in the full production table for REV_24_Q1.
inventory_v1.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Data sources are often undocumented, stale, or lack clear ownership, leading to wasted effort and flawed analysis. This skill automates the inventorying and profiling of supplied datasets to determine their suitability for specific analytical needs.
What it does
- Inventory sources builds a structured register of tables, APIs, files, and streams with metadata.
- Profile samples calculates completeness, validity, and distinct counts for supplied records.
- Freshness monitoring identifies stale tables based on expected refresh intervals.
- Governance mapping tracks data owners, access status, and organizational approvals.
- Suitability scoring provides a verdict on whether a source meets the requirements for a planned analysis.
How it works
- The user provides source metadata, access policies, schemas, and record samples.
- The agent generates a
request.jsonfollowing a strict schema and executes a deterministic Python script. - The script profiles the data and returns a structured result envelope.
- The agent generates a comprehensive report with an inventory, scorecard, and suitability verdict.
Frameworks & tools
This tool relies on a deterministic Python engine (scripts/run.py) and JSON schema validation. It operates locally on provided data samples and does not require live database connections.
Why this beats prompting it yourself
Prompting alone often leads to hallucinated statistics and inconsistent freshness logic. This skill uses a hardcoded execution script to ensure every count, ratio, and rule check is calculated correctly and traceably.
Use cases
- Auditing a data lake to find stale tables and missing owners.
- Profiling a new CSV or API response to verify if it contains enough valid data for a report.
- Assessing multiple datasets against a business glossary to see which one covers the required terms.
Known limitations
Profiles describe samples only and do not imply population values. It cannot connect to live systems or pull records directly; it only processes what you provide.
Profiles describe samples only. No live database connection. No data cleaning or fixing. Scope is limited to supplied metadata and record samples.
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