mshadmanrahman

search-first

Research existing solutions before writing custom code

security-review

Security checklist for code changes with severity classification

tdd-workflow

Test-driven development enforcement with RED-GREEN-IMPROVE cycle

verification-loop

Pre-commit verification with lint, type-check, tests, and security scan

strategic-compact

Monitor context usage and suggest compaction at logical milestones

handoff-doc

Create portable handoff document when switching sessions

meta-observer

Track skill performance and emerging patterns

orchestrator

Wave-based parallel task execution with sub-agents

session-init

Lightweight session initialization with context recovery

Enterprise knowledge assistant template. Searches Confluence, Jira, Slack, and internal docs to answer questions about your company. Users customize the company name, MCP sources, and team structure. Triggers on: "ask [company]", "who owns", "which team", "how does [thing] work".

critique

Pressure-test any document for logic gaps, unsupported assumptions, missing perspectives, and completeness. Works on PRDs, briefs, strategy docs, proposals, and any PM artifact. Triggers on: "critique this", "pressure test", "poke holes", "review this doc", "what am I missing", "devil's advocate".

deep-context

Cross-channel research on any topic. Searches all connected MCP tools in parallel, then synthesizes into a comprehensive briefing with timeline, status, stakeholders, and sources. Triggers on: "deep context", "tell me everything about", "research X", "full context on".

dogfood

Systematic web app QA using Playwright browser automation. Tests across viewports, checks navigation flows, form submissions, error states, accessibility basics, and visual consistency. Produces a structured bug report. Triggers on: "dogfood", "QA", "find bugs", "test this app".

issue-tree

Structure a messy problem into a MECE issue tree before diagnosing, planning, or deciding. Three tree types: Why (find causes), What (break down a deliverable into work), How (rank possible actions). Prevents mixing cause-finding, planning, and solutioning into one muddled discussion. Triggers on: "issue tree", "why tree", "root cause tree", "break this down", "structure this problem", "MECE", "what's actually going on with X".

lenny-podcast

Product management advisor drawing from 269+ Lenny's Podcast episodes and established PM frameworks. Searches transcripts for practitioner wisdom on strategy, growth, prioritization, user research, metrics, and product decisions. Triggers on: "what does Lenny say about", "ask lenny", "PM advice", "product strategy advice", "lenny podcast".

market-sizing

TAM/SAM/SOM market sizing analysis. Uses web search and research tools to produce a structured analysis with data sources, assumptions, calculations, and confidence ranges. Triggers on: "calculate TAM", "size the market", "market opportunity".

meeting-prep

Pre-meeting context gathering across all connected systems. Pulls recent interactions, open items, and suggested talking points for any person or topic. Triggers on: "meeting prep", "prep for my meeting", "meeting with X", "prep for 1:1".

people-sync

Pulls a Granola meeting transcript and updates people files with per-person intelligence: what they said, what they pushed back on, what they committed to. Compounds over time — each meeting adds a layer. Triggers on: "sync people", "update people from meeting", "update stakeholders", "/people-sync".

prd

Write product requirements documents at variable depth: one-pager, brief, full PRD, or RFC. Auto-detects depth from context. Reads existing research and context if available. Triggers on: "write a PRD", "product spec", "one-pager", "RFC", "requirements doc", "spec for", "brief for".

prioritize

Feature and initiative prioritization using 7 frameworks (RICE, ICE, WSJF, MoSCoW, Value/Effort, Kano, Weighted Scoring). Auto-detects mode: stack rank, scoring, opportunity assessment, trade-off analysis, or scope cut. Triggers on: "prioritize", "rank these features", "RICE score", "what should we build next", "scope cut", "trade-off".

synthesize-interviews

Turn raw user interviews, transcripts, and feedback into structured themes, pain points, and recommendations. Produces both a synthesis report and a standalone pain points document. Triggers on: "synthesize interviews", "interview findings", "user research synthesis", "what did users say", "analyze feedback".

weekly-status

Weekly accomplishment report. Pulls from Jira, GitHub, Confluence, and Slack to produce a structured status report with accomplishments, metrics, blockers, and next week plan. Triggers on: "weekly status", "status report", "what did I do this week".

product-discovery

Runs a full product discovery cycle for PMs with no UX researcher: frames the problem, plans Mom Test interviews, synthesizes findings into an opportunity tree, and packages evidence into a stakeholder one-pager. Use when the user wants to start discovery, validate a problem before building, plan customer interviews, turn interview notes into themes, or present research evidence to stakeholders. Triggers on "start product discovery", "validate this problem", "plan customer interviews", "synthesize my research", "package this for stakeholders".

design-review

Critique a design on layout, accessibility, responsiveness, dark mode and edge cases, with Pass, Needs work or Fail verdicts. Use on a built screen.

design-tokens

Establish color, typography, spacing, radius and motion as named role-based tokens. Use before building any interface.

grill-me

Two modes, picked from the file you point it at. Contract mode grills a claude-contract.md line by line until every line is a thing Claude must always do or must never do, editing that file in place and creating nothing new. Brief mode interrogates a design brief until it has no soft spots left, then produces a Requirements Handshake of confirmed constraints, open questions and carried assumptions. Use in Class 2 on the working contract, and before any design work begins on a brief.

heuristic-evaluation

Audit a design against Nielsen's ten usability heuristics, tying every finding to a specific element with a specific fix. Use on a built screen or a detailed mockup.

information-architecture

Map the user journey first, then derive the screen inventory, navigation and hierarchy from it, flagging any screen that serves no journey step. Writes the four-part markdown plus an HTML diagram of the step-to-screen mapping. Use after the brief is settled and before tokens or UI.

persona-acid-test

Stress-test a design through three lenses, confused user, skeptical engineer and impatient PM, before it goes to a stakeholder. Requires an actual design, an HTML file, Figma link or screenshot.

stuck

Diagnoses a broken course setup by reading the folder, so the student never has to describe the problem in English. Takes no argument. Reports which class they are on and which files are filled, then names one blocker and one fix: Claude Code opened at the wrong level, .claude/skills/ missing, skills not loading because the trust question was never answered, a file whose scaffolding was deleted before it was filled, work going into an answer-key .example.md, an unfilled YOUR TURN, a skill run before the files it reads were filled, or a hunt for a file no skill creates. When it cannot tell, it writes the group-chat message for the student. Use whenever a student says they are stuck, confused, blocked, or that something is not working.

manifest-reader

Read and summarize sub-agent manifest files

market-research

Competitive analysis and market research with source attribution. Parallel fan-out across web, review sites, and news. Triggers on: "market research", "competitive analysis", "competitor comparison".

writing-style

Voice profile configuration and application for consistent writing. Configure once with 2-3 samples, apply everywhere. Triggers on: "writing style", "voice profile", "apply my voice", writing long-form content.

writing-substack

Platform-optimized Substack Notes creation. 500-char limit, hook-driven, engagement-focused. Triggers on: "write a Substack Note", "Substack content", "Note ideas".

configure-pm-pilot

Interactive setup wizard for PM Pilot. Detects MCP tools, sets company name, creates memory template, validates installation.

shepherd-learn

Record why a triage verdict was wrong so future runs improve, building a learning log over time. Use after discovering a misclassified bug.

shepherd-review

Review triage results before syncing, so a human confirms verdicts on bugs that will be closed or reprioritized. Use between triage and sync.

shepherd-start

Set up Bug Shepherd for a project: create the triage config, identify the tracker, and establish the backlog baseline. Use when starting bug triage on a new project.

shepherd-sync

Sync triage results back into Jira, Linear, or GitHub Issues using the matching tracker adapter. Use after a triage run to update the tracker.

shepherd-triage

Launch parallel agents to check whether old bugs still reproduce on the live site, turning a stale backlog into a prioritized verified list. Use when a bug backlog needs cleaning without reading code.

crucible

Stress-test an idea under pressure by convening a multi-persona council that debates, rebuts, and returns a verdict. Use for heavy-lift decisions like moonlight strategy, product direction, career moves, or any question where the cost of being wrong is high. Triggers on "crucible", "/crucible", "run this through the crucible", "convene the crucible", "put X in the crucible", and Solo invocations like "give me the First Principles read on X". Not for tactical calls.

interview-snapshot

Transform a single customer interview transcript into a structured Interview Snapshot with five fixed sections. Use when the user has a Granola/Fireflies/manual transcript and needs the atomic research unit that feeds cross-interview synthesis. Triggers on "create interview snapshot", "process this interview", "synthesize this one interview", "turn this transcript into a snapshot". Prerequisite for any cross-interview synthesis step in discovery-process.

tech-to-pm-translator

Convert technical developer documentation into PM/designer-friendly context documents. Reads engineering docs (architecture refs, API docs, runbooks, ADRs, READMEs) and produces structured, code-free knowledge base files that non-engineers can use to understand systems, file better bugs, write better specs, and have informed conversations with engineers. Triggers on: "translate docs for PMs", "make this PM-friendly", "convert tech docs", "create PM context", "non-engineer version", "explain this for PMs", "designer-friendly docs", "/tech-to-pm".

brief-to-tasks

Break a design brief into executable time-boxed tasks, each with a done-when line. Use immediately before a build session.

design-brief

Turn an interrogated brief into the single source of truth for a design project. Use after /grill-me, before any IA or UI work.

frontend-design

Build an interface from the brief and the tokens rather than from a guess. Use only after grill-me, design-brief, information-architecture and design-tokens have run.