kimsanguine

79 mods in 6 plugins, 3 repositories, 70 stars between them.

kimsanguine/AI_PM

Skill Claude CodeCodex

A tool that compares three to five competitors using the same criteria and creates a comparison table and battlecard. A battlecard is a short reference for explaining how a product differs from competitors.

not rated 57 2mo ago A 58 tokens

kimsanguine/AI_PM

Skill Claude CodeCodex

A tool that combines interview transcripts, survey responses, and product-usage logs into an insights table for PM decisions. Low-frequency findings are labelled as weak signals instead of being treated as confirmed insights.

not rated 57 2mo ago A 65 tokens

experiment-analyzer

03

kimsanguine/AI_PM

Skill Claude CodeCodex

A tool that reads A/B test results from CSV files or tables and writes a report section for a PM. An A/B test compares two versions, while guardrails are safety or quality limits that results must not break.

not rated 57 2mo ago A 55 tokens

jtbd

04

kimsanguine/AI_PM

Skill Claude CodeCodex

A method for turning interviews and feature requests into job statements: what situation a person is in, what they want to do, and what result they expect. A job is the task someone is trying to accomplish, not the feature they ask for.

not rated 57 2mo ago A 88 tokens

kpi-card-builder

05

kimsanguine/AI_PM

Skill Claude CodeCodex

A builder for KPI definition cards. A KPI, or key performance indicator, is a measurement used to track a goal; each card records its name, formula, data source, response threshold, owner, and counterexamples.

not rated 57 2mo ago A 67 tokens

kimsanguine/AI_PM

Skill Claude CodeCodex

A planning tree that starts with a desired outcome and connects user needs to possible solutions and tests. An opportunity means a user need or pain, not a proposed solution.

not rated 57 2mo ago A 82 tokens

prd-generator

07

kimsanguine/AI_PM

Skill Claude CodeCodex

A guided tool for creating a product requirements document, or PRD—the written plan for what a product or feature should do. It uses Socratic questions to challenge assumptions and presents multiple options.

not rated 57 2mo ago A 40 tokens

working-backwards

08

kimsanguine/AI_PM

Skill Claude CodeCodex

A product-planning method that starts by writing the launch press release and frequently asked questions before building the product. The documents describe customer value and answer difficult questions in plain language.

not rated 57 2mo ago A 72 tokens

briefing

09

kimsanguine/AI_PM

Command Claude Code

A command that gathers recent signals from Slack, Linear, Notion, GitHub, and a calendar, then condenses them into five priorities for a product manager’s day.

not rated 57 2mo ago A 0 tokens

competitor

10

kimsanguine/AI_PM

Command Claude Code

A command for comparing three to five named competitors using the same criteria and producing a structured analysis table plus a draft battlecard.

not rated 57 2mo ago A 0 tokens

discovery

11

kimsanguine/AI_PM

Command Claude Code

An analysis command that combines interview transcripts, survey results, and usage logs such as CSV, JSON, or notes. It produces an insight table that a product manager can turn into a product requirements document or experiment.

not rated 57 2mo ago A 0 tokens

pm-setup

12

kimsanguine/AI_PM

Command Claude Code

A project setup command for product managers who are not developers and are starting to use Claude Code.

not rated 57 2mo ago A 0 tokens

postmortem

13

kimsanguine/AI_PM

Command Claude Code

A command for writing a blameless postmortem, a review of a launch, experiment, or incident that focuses on events and causes rather than assigning fault.

not rated 57 2mo ago A 0 tokens

prd

14

kimsanguine/AI_PM

Command Claude Code

A command that creates a first draft of a PRD, or product requirements document describing what a new feature or project should achieve.

not rated 57 2mo ago A 0 tokens

retro

15

kimsanguine/AI_PM

Command Claude Code

A retrospective command that reviews recent Claude Code conversations for repeated fixes, instructions, and mistakes.

not rated 57 2mo ago A 0 tokens

review

16

kimsanguine/AI_PM

Command Claude Code

A review command for checking a pull request, chapter, or document through four verification stages.

not rated 57 2mo ago A 0 tokens

start-1

17

kimsanguine/AI_PM

Command Claude Code

A guided beginner course for product managers learning Claude Code, a coding assistant that works from the command line.

not rated 57 2mo ago A 0 tokens

start-2

18

kimsanguine/AI_PM

Command Claude Code

A guided course for product managers who already know basic conversations with Claude Code.

not rated 57 2mo ago A 0 tokens

start-3

19

kimsanguine/AI_PM

Command Claude Code

A guided product-management exercise covering Discovery, Definition, and Delivery, from understanding a problem to handing over a solution.

not rated 57 2mo ago A 0 tokens

start

20

kimsanguine/AI_PM

Command Claude Code

A guided course for learning Claude Code through hands-on exercises rather than reading alone.

not rated 57 2mo ago A 0 tokens

status

21

kimsanguine/AI_PM

Command Claude Code

A weekly product-management status report command that summarizes work for the week.

not rated 57 2mo ago A 0 tokens

today

22

kimsanguine/AI_PM

Command Claude Code

A daily product-management briefing command that organizes the day's urgent and ongoing work.

not rated 57 2mo ago A 0 tokens

llm-brain

23

kimsanguine/llm-brain

Plugin Claude Code

Bundles 1 skill, 7 commands · 184 tokens together

A personal knowledge system that turns raw notes into an organised, searchable wiki and can export them in an open format. It also keeps separate layers for tasks, experiences, concepts, procedures and system information.

not rated 10 16d ago A tokens not measured original MIT

architect

24

kimsanguine/hplan

Plugin Claude Code

Bundles 4 skills · 259 tokens together

A set of guidance for designing AI-agent systems: software where several AI workers may coordinate to complete a task. It covers coordination patterns, business models, competitive advantages, and memory planning.

not rated 2 2mo ago A tokens not measured original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: