Agent skill

customer-feedback-analysis

Extract themes from customer feedback (JTBD interviews, churn surveys, G2 reviews, support tickets) with verbatim discipline and route per theme class.

slowingNeeds a keyActs undeclared505 words

Filed under Onboarding, retention and expansion.

From devangk003/gtm-agent-skills · 32 skills · 0 · pushed 2026-06-18

What it does when it runs

Extract themes from customer feedback (JTBD interviews, churn surveys, G2 reviews, support tickets) with verbatim discipline and route per theme class. Use when the user says "analyze customer interviews", "churn survey themes", "extract JTBD themes", or "product feedback patterns.

Read from the skill and the 4 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply: AGENTIC_APP_TOKEN
LINEAR_API_KEY
Hosts it reaches: No third-party host appears in the skill or its bundled files.
Tool permissions it declares: No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files: shell

Ask about customer-feedback-analysis

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?Claude ChatGPT

Adapt it to my stackClaude ChatGPT

What else do I need for it to workClaude ChatGPT

Rather ask a human? Talk to Cheetah

Install it

View source on GitHub ↗

git clone --depth 1 --filter=blob:none --sparse https://github.com/devangk003/gtm-agent-skills.git /tmp/gtm-agent-skills
git -C /tmp/gtm-agent-skills sparse-checkout set "customer-feedback-analysis"
mkdir -p ~/.claude/skills/customer-feedback-analysis
cp -R "/tmp/gtm-agent-skills/customer-feedback-analysis/." ~/.claude/skills/customer-feedback-analysis/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Before you install: this skill will not complete its job on a bare agent. It needs AGENTIC_APP_TOKEN, LINEAR_API_KEY, which you have to obtain separately.

The skill

Source on GitHub ↗

Reproduced in full from devangk003/gtm-agent-skills/blob/6b9a1b0094ffd83d6c02bc03b7ce1610661a1644/customer-feedback-analysis/SKILL.md, which is licensed MIT (skill frontmatter). 505 words, 6 headings.

Customer Feedback Analysis

Extract structured themes from unstructured customer feedback using qualitative coding, aggregate sentiment + frequency patterns, and route per theme class. Hard rule: every extracted theme references a verbatim customer quote — never paraphrase as the primary record.

When to Use

≥5 won-deal JTBD interviews completed — extract themes
Churn survey responses accumulated — analyze
G2/Capterra reviews landed — what are customers saying
Support ticket themes — feed product priorities
Conversation intelligence flagged feature-request pattern — confirm + route
User says "analyze customer interviews" or "churn survey themes"

Quick Reference

Theme class	Detection signal	Default routing
Pain-job-anchor	Underlying job they hired product for (JTBD)	→ icp-definition + positioning-strategy
Forces-of-progress (push/pull/habit/anxiety)	Why they switched / what held them back / what worried them	→ positioning-strategy + icp-refinement-loop
Positive outcome	"Saved X hours" / "ramp dropped"	→ kpi-reporting + case-study source
Feature request	"Wish you had X" / "we worked around X"	→ product backlog
Bug/friction	"Broke when X" / "couldn't figure out Y"	→ product backlog priority
Pricing reaction	"Worth it" / "expensive at X"	→ revenue-forecasting
Competitor mention	Named competitor referenced	→ competitive-intelligence
Champion language	What champion said internally to win deal	→ cold-email-sequence copy bank

Concept	Value
Min corpus per source	5 items for aggregation; below → individual items only
JTBD methodology	Bob Moesta — 4-forces (push/pull/habit/anxiety)
ICP-implication threshold	Theme in ≥30% of won-deal interviews → icp-refinement-loop flag

Procedure

Validate inputs + classify per item. Pull corpus per source. Min ≥5 for aggregation. LLM-backed per-item classification against 8-class taxonomy. See ${HERMES_SKILL_DIR}/references/theme-taxonomy.md.
Aggregate + analyze. Group by theme class + entity; count frequency; segment by cohort/tier. Per theme: sentiment distribution + frequency vs threshold + segment concentration.
Route per theme class. Pain → icp-definition. Forces → positioning-strategy. Feature → product backlog. Pricing → revenue-forecasting. Competitor → competitive-intelligence. Champion → copy bank. See ${HERMES_SKILL_DIR}/references/routing-map.md.
ICP-implication check. Theme ≥30% in won-deal segment → flag for icp-refinement-loop.
Push + compose report. Per-theme records with verbatim + provenance. PATCH person/company tags with feedback signals. See ${HERMES_SKILL_DIR}/scripts/push_to_crm.py.

Pitfalls

Paraphrasing instead of verbatim — themes need actual customer language
Aggregating below threshold — 4 churn responses don't make a pattern
JTBD without 4-forces frame — misses anxieties + habit; use Moesta's push/pull/habit/anxiety
Missing champion-language goldmine — won-deal interviews surface pure copy material
Multi-language handled in English only — source-language extraction + English summary
Routing without frequency — single request is noise; aggregate first

Verification

Every extracted theme cites a verbatim quote + source
Aggregation thresholds met before pattern-claims
ICP-implication flags trace to ≥30% threshold
Product backlog routes (when configured) created actual issues

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

customer-research by coreyhaines31 · 45,947
yt-competitive-analysis by ericosiu · 3,449
competitor-analysis by OpenClaudia · 664
content-gap-analysis by OpenClaudia · 664
geo-analysis by OpenClaudia · 664
video-ad-analysis by OpenClaudia · 664
customer-story-writer by zapier · 329
find-customer-social-proof by zapier · 329

Need help setting it up?

This page tells you what customer-feedback-analysis does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.