How it works

Inputs in, an ordered run, a result out.

A Sequential Agent takes typed inputs, runs its tasks top to bottom against your tools and tables, and returns a structured result, the same way on every run.

Typed inputs

Values you pass in, validated before the run spends anything.

2 tasks

Each task runs in order, passing its output to the next.

Runs standalone

No external tools or tables needed.

Result + trace

A structured output, with every step recorded.

The run loop

Read→Plan→Act→Check→Write→Tracerepeats until done

What's inside

Everything this template installs, by name.

Tools & tables

Runs standalone, no external tools or tables required

Tasks

2 ordered tasks

What it takes in

The inputs you provide on each run, validated before anything executes.

InputTypeRequired

Feedback DumpfeedbackDumpLong textRequired

Product ContextproductContextLong textOptional

The work, step by step

2 tasks, run in order.

Each task passes its result to the next. Here's what each does and what it produces.

01
Cluster the feedback. Read every item in {{feedbackDump}} and group the feedback into distinct themes. For each theme give a clear name, a one-sentence summary of what users are saying, a count of how many feedback items support it, and 1 to 3 representative verbatim quotes copied exactly from the dump: never invent, paraphrase, or embellish quotes. If {{productContext}} is provided, use it to interpret ambiguous feedback correctly. Put items that fit no theme into a Miscellaneous bucket rather than forcing them into a cluster. Present readably, never JSON.
Expected output
A readable set of Feedback Themes with no JSON and no code blocks. For each theme, a heading with: Theme Name, Summary (one sentence), Count (number of supporting feedback items), and Representative Quotes (1 to 3 exact verbatim quotes in quotation marks). End with a Miscellaneous section listing unclustered items, or a note that there were none.
02
Write the insight report. Using the themes from Task 1, produce a prioritized insight report. Rank the themes by frequency combined with severity (how much pain the theme represents, judged from the language and stakes in the quotes) and give a one-line rationale for each ranking. Split recommendations into Quick Wins (high impact, low effort) and Strategic Bets (high impact, larger investment), each with a concrete recommended next action and the theme it addresses. Close with a Not Worth Acting On section listing themes or requests to consciously deprioritize and why. Ground every claim in the clustered feedback; if {{productContext}} is provided, align recommendations with it. Present readably, never JSON.
Expected output
A readable insight report with no JSON and no code blocks, using these headed sections: Prioritized Themes (ranked list, each with a one-line frequency-and-severity rationale), Quick Wins (bullets, each with a concrete next action and the theme it addresses), Strategic Bets (bullets, each with a concrete next action and the theme it addresses), and Not Worth Acting On (bullets, each with the reason to deprioritize).

Related templates

Sequential Agent

PRD Drafter

Turns a feature idea into a complete, review-ready PRD. Distills the problem, target users, goals, and success metrics, drafts a full PRD with user stories, acceptance criteria, functional requirements, edge cases, open questions, and a launch checklist, then red-teams it with the five hardest reviewer questions. Pure standalone: paste the idea, get a PRD you can circulate today. No tools or tables required.

Sequential Agent

Competitive Teardown

Produces a structured competitive teardown and battle-ready positioning from pasted competitor info. Breaks down the competitor's positioning, ICP, pricing model, claimed strengths, and visible weaknesses, runs an honest head-to-head against your product, then arms sales with talking points, landmine questions, and a one-paragraph positioning statement. Pure standalone: paste competitor material and your product details. No tools or tables required.