- Home
- Skills
- Social Media
- Social Metrics Preflight
Works with the AI Agents you already use
Social Metrics Preflight
Validate social media metrics exports for schema errors, duplicates, and formatting issues before analysis.
Free
Social Metrics Preflight
Example session with this skill installed
Check this metrics export for errors: date,platform,metric,value,source
2024-01-01,X,Reach,5000,native
01/02/2024,X,Reach,1.2k,
2024-01-01,X,Reach,5000,native
2024-01-03,IG,Likes,-5,native
- Read your context and instructions
- Compiled the social metrics preflight
- Generated the data export
I have completed the validation check on the provided metrics data. The export contains critical formatting and data integrity issues that must be addressed before analysis.
Metrics preflight — metrics_export.csv
Verdict: FIX · 1 BLOCK · 2 FIX · 1 PASS
BLOCK
| Item | Evidence | Fix |
|---|---|---|
| Values not numeric | row 2: "1.2k" | store as 1200 |
| Negative values | row 4: "-5" | verify and correct negative reach/likes |
FIX
| Item | Evidence | Fix |
|---|---|---|
| Missing source | row 2 | add the export source and date |
| Mixed date format | row 2: "01/02/2024" vs "2024-01-01" | use ISO dates (YYYY-MM-DD) |
| Duplicate rows | row 1 and 3 identical | dedupe |
PASS
Required columns (date, platform, metric, value, source) are all present in the header.
Next steps
- Convert "1.2k" to numeric 1200 and fix the negative value in row 4.
- Standardize all dates to ISO format and remove the duplicate entry in row 3.
- Once clean, use
social-competitor-metrics-auditorfor the platform comparison.
Nothing was changed. No values were invented.
metrics_export.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
Corrupt or inconsistent metrics data will break your dashboards and lead to false insights. This skill acts as a strict validator for social media and competitor metrics exports, identifying structural flaws, data type errors, and missing provenance before you waste time on analysis.
What it does
- Schema validation checks for required columns like
date,platform,metric,value, andsource. - Row hygiene audits identifies blank cells, duplicate entries, and impossible values like negative reach.
- Format enforcement flags mixed date formats, units stored as text, and numeric outliers.
- Provenance tracking ensures every data point has a verified source or platform tag.
- Categorized reporting issues a clear
PASS,FIX, orBLOCKverdict with specific row evidence.
How it works
- Context gathering requires the file content or a sample of the columns and rows for inspection.
- Execution workflow runs the data through schema, hygiene, and range checks.
- Verdict generation produces a structured report quoting offending rows and required fixes.
Frameworks & tools
This tool works with CSV, TSV, and Markdown tables exported from platforms like Sprout Social, Hootsuite, or native social analytics.
Why this beats prompting it yourself
Standard LLM prompts often hallucinate missing values or silently ignore formatting errors. This skill follows a rigid validation logic that refuses to invent data and forces a formal BLOCK or FIX status, ensuring zero-trust data integrity.
Use cases
- Validating a messy CSV export before importing it into Tableau or PowerBI.
- Checking for duplicate rows and text-formatted numbers in a pasted spreadsheet.
- Identifying missing source tags in a multi-platform competitor audit.
Known limitations
This is a read-only validator. It will not modify your files, fill in missing values, or perform cross-platform metric comparisons.
Read-only validation only. Does not modify files, fill gaps, or analyze trends. Requires CSV, TSV, or table data. Use social-competitor-metrics-auditor for methodology.
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
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 26 days ago
- Free with an Agensi account
Permissions
File Scopes