analyze-sequences — Claude Code command: what it does, how to install
analyze-sequences reviews a running or finished email sequence and reports performance, deliverability and test results, with a prioritized action plan.
Apache-2.0 ★ 450 updated 2026-04-03
.claude/commands/analyze-sequences.mdWhat analyze-sequences does
The command runs a report on an email sequence: it pulls metrics from the MAP and the CRM for the window you give, checks KPI attainment, and looks for drop-off points, inbox placement problems and signs of list fatigue. It handles sequences that are still live and ones already closed.
Scope comes from flags. --sequence names the program or its ID, --window takes 7d, 14d, 30d or a custom date range, --detail picks summary or full, --segment narrows to a persona, tier or region, and --compare_to sets a baseline sequence or control group. After that comes the experiment readout with confidence intervals and a prioritized plan: adjust cadence, refresh copy, escalate what broke.
You get a dashboard snapshot with KPIs split by touch, persona and mailbox provider, an alert summary for deliverability that points at deliverability-lead incident logs, and a memo on the experiments. Both the MAP and the CRM feed those numbers, so a gap in either system leaves holes in the readout. Work is partly handed to the experiment-analyst and deliverability-lead agents and the cadence-design skill.
What it helps with
- Find the step where a sequence loses replies
- Trace an engagement dip back to a mailbox provider
- Close out a running test and decide what to roll out
- Catch list fatigue before the numbers drop
- Hand architects and ops a recap with backlog updates
Example prompts for the agent
- Run
/email-sequence-orchestration:analyze-sequences --sequence "Expansion Play" --window 14d --detail fulland walk me through the whole readout. - Compare the onboarding sequence against control over the last 30 days:
/email-sequence-orchestration:analyze-sequences --sequence "Onboarding" --window 30d --compare_to "Control" --detail summary. - Check deliverability and fatigue for the Enterprise segment:
/email-sequence-orchestration:analyze-sequences --sequence "Expansion Play" --window 7d --segment "Enterprise" --detail summary.
How to install analyze-sequences
Manually: put the files in place
.claude/commands/analyze-sequences.mdCall it with /analyze-sequences. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/analyze-sequences.mdCall it with /prompts:analyze-sequences. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/analyze-sequences.prompt.mdManually: put the files in place
.gemini/commands/analyze-sequences.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about analyze-sequences
How do I install analyze-sequences in Claude Code?
Copy the command file to .claude/commands/analyze-sequences.md in your project. Claude picks it up there, and you call it as /email-sequence-orchestration:analyze-sequences.
What does analyze-sequences do?
It gathers sequence metrics from the MAP and CRM, checks deliverability and drop-off points, reads out experiments, and returns a prioritized list of actions.
What data does it need to work?
MAP and CRM metrics for the chosen window. Set the window, audience and baseline with --window, --segment and --compare_to.
Source text
analyze-sequences.md · Apache-2.0
--- name: analyze-sequences description: Evaluates live or completed email sequences, highlighting performance, deliverability, and experiment results. usage: /email-sequence-orchestration:analyze-sequences --sequence "Expansion Play" --window 14d --detail full --- # Command: analyze-sequences ## Inputs - **sequence** – program name or ID to analyze. - **window** – time horizon (7d, 14d, 30d, custom date range). - **detail** – summary | full to control verbosity. - **segment** – optional audience filter (persona, tier, region). - **compare_to** – optional baseline sequence or control group. ## Workflow 1. **Data Pull** – collect MAP + CRM metrics (deliverability, engagement, pipeline impact) within specified window. 2. **Health Diagnostics** – evaluate KPI attainment, drop-off points, inbox placement, and fatigue risk. 3. **Experiment Readout** – summarize running/closed tests with confidence intervals and recommended actions. 4. **Insights & Actions** – generate prioritized playbook (optimize cadence, refresh content, escalate issues). 5. **Handoff** – prepare recap doc plus suggested backlog updates for architect + ops teams. ## Outputs - Performance dashboard snapshot with KPIs by touch, persona, mailbox provider. - Deliverability + QA alert summary referencing `deliverability-lead` incident logs. - Experiment insights memo (winners, next tests, rollout guidance). ## Agent/Skill Invocations - `experiment-analyst` – interprets tests and recommends next experiments. - `deliverability-lead` – correlates reputation signals with performance dips. - `cadence-design` skill – flags pacing issues causing fatigue. ---
© gtmagents · Apache-2.0 · gtmagents/gtm-agents