- Home
- Skills
- Business & Operations
- 🩹 Customer Pain Extractor
More screenshots
Works with the AI Agents you already use
🩹 Customer Pain Extractor
Flag and group keyword-based candidate pain signals in supplied feedback, with quotes and text-occurrence counts for human review. Select CSV feedback columns explicitly. Counts are not respondents or proof of demand.
$12
🩹 Customer Pain Extractor
Example session with this skill installed
Run the bundled synthetic fixtures/risky-sample/feedback.txt and show the cue counts and quotes. Do not interpret the counts as people or demand.
- Read your context and instructions
- Compiled the customer pain extractor
Actual synthetic fixture run
This example uses invented feedback bundled with the package, not customer evidence.
# Customer pain extractor
Found 5 matched statements with 7 rule matches.
Counts describe supplied text occurrences, not unique customers, buyers or proven demand.
## 1. switching risk (2 mentions, score 3)
- "I cancelled because the import was broken every time."
- "We found a better tool after the app stayed too slow."
## 2. ease of use (1 mentions, score 1)
- "The setup is confusing and I can't tell what to do next."
## 3. missing feature (1 mentions, score 1)
- "I wish it had a simple weekly report for my team."
## 4. price and value (1 mentions, score 1)
- "The pricing feels too expensive for a tiny project."
Review the original context before accepting any cue as a complaint. These counts do not establish five respondents, willingness to pay or a market opportunity. Scores are within-sentence keyword weights, not a 1–10 severity rating.
Connects securely to your tools. The creator never sees your data.
About this skill
Review feedback with traceable cue counts
Customer Pain Extractor reads authorized local text, Markdown or CSV feedback and flags configured phrases associated with frustration, wishes, switching and price. It groups matched text by configured themes and returns quotes, rule-match counts and matched-statement counts. One statement may match multiple rules; repeated text remains repeated occurrences.
What you get
A standard-library Python scanner, editable cue configuration, agent workflow, interpretation checklist and synthetic regression tests. JSON includes source paths and files-scanned coverage. Markdown presents grouped example quotes. These outputs help organize review; they do not establish genuine complaints, unique respondents, buying intent or demand.
Select the right input
For survey exports, identify the exact feedback headers and use --csv-column for each. Headers and unselected metadata are excluded. Without column selection, CSV is treated as headerless feedback-only text. UTF-8 and BOM-marked UTF-16 are supported. Missing or unreadable inputs, malformed or oversized CSV fields and folders with no supported files return no assessment. Linked and reparse-point inputs are refused.
Boundaries
Requires Python 3.10+ and an agent with authorized local execution. No scraping, network access or input-file writes. Matching is heuristic: negation, substrings, paraphrases and punctuation can cause false positives or misses. Scores count configured intensity words, not business priority. Confirm each quote before suggesting hypotheses or next steps. Price and existing terms remain unchanged.
Keyword matching is not semantic complaint detection. Negations, substrings and punctuation can produce false positives or misses. Counts are text occurrences/rule matches, not people or demand. Intensity scores are keyword weights. Select CSV feedback columns explicitly; the header must be the first line. Duplicate selections, missing columns and unsupported inputs are refused. Generic errors require checking input format/permissions. Extreme directory depth, resource exhaustion or interrupted output is not a completed assessment. Windows link tests were limited by local privileges; independent link/permission tests ran on Linux.
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 16 days ago
Requires
Permissions
File Scopes
Reads authorized local feedback plus bundled cue configuration. No network, credentials, environment-variable access or input-file writes. Invalid or unreadable input aborts assessment; linked/reparse inputs are refused.