measure-engagement — Claude Code command for community health metrics
measure-engagement pulls community health metrics, sentiment, and experiment results into one report with a prioritized action list. Use it when you need to see what is happening in a community and what to do about it.
Apache-2.0 ★ 450 updated 2026-04-03
.claude/commands/measure-engagement.mdWhat measure-engagement does
The command runs a pipeline: it pulls platform analytics and CRM attribution, computes growth and engagement metrics by dimension, scans conversations and NPS for emerging themes, evaluates running pilots against guardrails, and produces a prioritized action list with owners and deadlines.
The plan is stored as JSON in .claude/plans/plan-<timestamp>.json, which keeps an audit trail. Failed data pulls retry twice before escalation along the Community Analyst → Community Lead → CS/Product chain. The workflow can branch into a diamond when diagnostics and experiment analysis run in parallel.
You invoke it with /community-orchestration:measure-engagement plus a time window (--window 7d|30d|90d) and report depth. The report references findings from the sentiment-analysis skill. Without CRM and platform analytics data, the output will be incomplete.
What it helps with
- pull platform and CRM metrics into one snapshot
- surface themes and risks from community conversations
- check running pilots against KPIs and guardrails
- get an action list with owners and deadlines
- keep a plan with dependencies for audit
Example prompts for the agent
- Run
/community-orchestration:measure-engagement --window 30d --detail full --dimensions "channel,persona" - Review engagement for the last 90 days and break it down by program
- Build a readout for the current experiment and flag anything off KPI
How to install measure-engagement
Manually: put the files in place
.claude/commands/measure-engagement.mdCall it with /measure-engagement. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/measure-engagement.mdCall it with /prompts:measure-engagement. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/measure-engagement.prompt.mdManually: put the files in place
.gemini/commands/measure-engagement.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about measure-engagement
How do I install measure-engagement in Claude Code?
Place the command file at .claude/commands/measure-engagement.md. After that, invoke it as /community-orchestration:measure-engagement.
What does measure-engagement do?
It aggregates platform analytics, CRM attribution, and sentiment data, computes community health metrics, and outputs a prioritized action list with owners and deadlines.
What parameters does the command accept?
Required inputs are --window (7d, 30d, or 90d) and --detail (summary or full). Optional flags include --dimensions for breakdowns by channel, persona, program, or cohort, plus experiment lists and an alert threshold.
Source text
measure-engagement.md · Apache-2.0
--- name: measure-engagement description: Reviews community health metrics, sentiment, and experiment impact with recommended actions. usage: /community-orchestration:measure-engagement --window 30d --detail full --dimensions "channel,persona" --- # Command: measure-engagement ## Inputs - **window** – time horizon (7d, 30d, 90d) for analysis. - **detail** – summary | full report depth. - **dimensions** – optional breakdown (channel, persona, program, cohort). - **experiments** – optional list of programs to analyze. - **alert_threshold** – optional metric threshold for escalations. ### GTM Agents Pattern & Plan Checklist > Mirrors GTM Agents orchestrator blueprint @puerto/plugins/orchestrator/README.md#112-325. - **Pattern selection**: Engagement measurement typically runs **pipeline** (data → diagnostics → sentiment → experiment readouts → actions). If diagnostics + experiment analysis can run in parallel, log a **diamond** segment with merge gate in the plan header. - **Plan schema**: Save `.claude/plans/plan-<timestamp>.json` capturing window, data feeds, task IDs, dependency graph (analytics, CRM, sentiment tools), error handling, and success metrics (engagement %, advocacy, risk volume). - **Tool hooks**: Reference `docs/gtm-essentials.md` stack—Serena for schema diffs, Context7 for platform documentation/conversation exports, Sequential Thinking for insights review cadence, Playwright for verifying dashboard/report embeds. - **Guardrails**: Default retry limit = 2 for failed data pulls or sentiment processing; escalation ladder = Community Analyst → Community Lead → CS/Product leadership. - **Review**: Run `docs/usage-guide.md#orchestration-best-practices-puerto-parity` before execution to confirm inputs, dependencies, deliverables. ## Workflow 1. **Data Pull** – aggregate platform analytics, CRM attribution, sentiment scores, and support signals. 2. **Health Diagnostics** – compute growth, activation, engagement, retention, and advocacy metrics by dimension. 3. **Sentiment Review** – scan community conversations, surveys, and NPS for emerging themes. 4. **Experiment Readouts** – evaluate running pilots against guardrails + KPIs. 5. **Action Recommendations** – produce prioritized playbook (content tweaks, ambassador outreach, escalations). ## Outputs - Community health dashboard snapshot with annotated insights. - Sentiment + risk report referencing `sentiment-analysis` findings. - Recommended action list with owners, deadlines, and expected impact. - Plan JSON entry stored/updated in `.claude/plans` for audit trail. ## Agent/Skill Invocations - `community-analyst` – leads data aggregation and insights. - `sentiment-analysis` skill – interprets qualitative signals and escalation needs. - `community-lead` – receives action plan and drives follow-through. ## GTM Agents Safeguards - **Fallback agents**: document substitutes (e.g., Community Lead covering Analyst) when analysts unavailable. - **Escalation triggers**: if engagement drops below alert_threshold twice or sentiment exposes severe risks, escalate via GTM Agents rip-cord to cross-functional leadership. - **Plan maintenance**: update plan JSON/change log when data sources, thresholds, or reporting cadences change to maintain GTM Agents audit parity. ---
© gtmagents · Apache-2.0 · gtmagents/gtm-agents