How do I install the Survey Thematic Analyzer skill?

Run npx skillmds@latest add themarmack/survey-thematic-analyzer in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Survey Thematic Analyzer safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Survey Thematic Analyzer?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Survey Thematic Analyzer free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Survey Thematic Analyzer?

themarmack (@themarmack) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Survey Thematic Analyzer by themarmack

Name: Survey Thematic Analyzer
Author: themarmack

Survey Thematic Analyzer

Cluster open-text survey responses into themes with representative quotes. Takes raw responses (developer experience surveys, Copilot rollout retros, post-incident surveys) and produces a structured analysis with theme labels, response counts per theme, 2-3 representative quotes per theme, and outlier responses worth surfacing individually. Use when analyzing developer surveys or Copilot rollout feedback.

themarmack 444982d 2.2 KB Updated Sep 17, 2026

File contents

survey-thematic-analyzer

The qualitative-analysis skill. Closed-form survey data is easy to analyze; open-text responses are where the real signal lives but the analysis cost has historically been prohibitive.

When to use

Developer experience survey (annual / pulse).
Copilot rollout retro responses.
Post-incident "what would have helped" survey.
Conference attendee feedback.

When NOT to use

Closed-form (yes/no, Likert) responses → standard quantitative tools.
Interviews → different methodology; use a transcript-specific tool.

Workflow

Read responses: structured list of {respondent_id, role, response_text}.
Cluster by theme using NLP grouping or LLM-driven cluster identification.
For each theme: label, count, 2-3 verbatim representative quotes (with respondent context).
Surface outliers: responses that don't cluster — sometimes the most informative.
Cross-reference vault: do any themes connect to known objections, decisions, or open questions?
Persist the analysis — after presenting it in chat, write the full analysis via vault-writer.write_insight to vault/insights/YYYY-MM-DD-survey-analysis-{slug}.md so themes stay queryable across survey cycles.

Output structure

# Survey Analysis — {survey name}

## Sample
- N responses
- Roles distribution: ...

## Themes
### Theme 1: {label} (N responses, X%)
- Representative quotes:
  - "{quote}" — {role}
- Vault connection: [[wikilink if applicable]]

### Outliers worth surfacing
- {quote} — {why it's worth noting}

Lands at vault/insights/YYYY-MM-DD-survey-analysis-{slug}.md, written via vault-writer.write_insight (workflow step 6).

Acceptance test

SKILL.md describes the workflow. Live exercise deferred to first actual survey input.

---name: survey-thematic-analyzerdescription: Cluster open-text survey responses into themes with representative quotes. Takes raw responses (developer experience surveys, Copilot rollout retros, post-incident surveys) and produces a structured analysis with theme labels, response counts per theme, 2-3 representative quotes per theme, and outlier responses worth surfacing individually. Use when analyzing developer surveys or Copilot rollout feedback.---# survey-thematic-analyzerThe qualitative-analysis skill. Closed-form survey data is easy to analyze; open-text responses are where the real signal lives but the analysis cost has historically been prohibitive.## When to use- Developer experience survey (annual / pulse).- Copilot rollout retro responses.- Post-incident "what would have helped" survey.- Conference attendee feedback.## When NOT to use- Closed-form (yes/no, Likert) responses → standard quantitative tools.- Interviews → different methodology; use a transcript-specific tool.## Workflow1. **Read responses**: structured list of `{respondent_id, role, response_text}`.2. **Cluster by theme** using NLP grouping or LLM-driven cluster identification.3. **For each theme**: label, count, 2-3 verbatim representative quotes (with respondent context).4. **Surface outliers**: responses that don't cluster — sometimes the most informative.5. **Cross-reference vault**: do any themes connect to known objections, decisions, or open questions?6. **Persist the analysis** — after presenting it in chat, write the full analysis via `vault-writer.write_insight` to `vault/insights/YYYY-MM-DD-survey-analysis-{slug}.md` so themes stay queryable across survey cycles.## Output structure```markdown# Survey Analysis — {survey name}## Sample- N responses- Roles distribution: ...## Themes### Theme 1: {label} (N responses, X%)- Representative quotes: - "{quote}" — {role}- Vault connection: [[wikilink if applicable]]### Outliers worth surfacing- {quote} — {why it's worth noting}```Lands at `vault/insights/YYYY-MM-DD-survey-analysis-{slug}.md`, written via `vault-writer.write_insight` (workflow step 6).## Acceptance testSKILL.md describes the workflow. Live exercise deferred to first actual survey input.

Frequently asked questions