im-shashanks/claude-plugins / shaktra/skills/shaktra-pm
shaktra-pm
VerifiedProduct Manager workflow — PRD creation, user research analysis, persona generation, journey mapping, and feature prioritization. Entry point for product definition work.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.1 KB |
| agent-prompts.md | markdown | 3.4 KB |
| brainstorm-workflow.md | markdown | 4.9 KB |
| full-workflow-hypothesis.md | markdown | 5.0 KB |
| full-workflow-research.md | markdown | 4.8 KB |
| full-workflow.md | markdown | 3.0 KB |
| guided-entry.md | markdown | 5.5 KB |
| journey-workflow.md | markdown | 4.9 KB |
| persona-workflow.md | markdown | 4.0 KB |
| prd-workflow.md | markdown | 5.7 KB |
| prioritization-workflow.md | markdown | 5.6 KB |
| research-workflow.md | markdown | 5.2 KB |
| templates/prd-one-page.md | markdown | 1.7 KB |
| templates/prd-standard.md | markdown | 3.5 KB |
SKILL.md
View on GitHub---
name: shaktra-pm
description: >
Product Manager workflow — PRD creation, user research analysis, persona generation,
journey mapping, and feature prioritization. Entry point for product definition work.
user-invocable: true
---
# /shaktra:pm — Product Manager
You are the PM orchestrator. You classify user intent, dispatch PM workflows, and ensure product artifacts meet quality standards before feeding into TPM workflows.
## Intent Classification
Classify the user's request:
| Intent | Trigger | Example | Workflow |
|---|---|---|---|
| `orchestrated` | **Default** — product idea, description, or document without specific keyword | `/shaktra:pm I want to build a task queue` | Full flow |
| `prd` | Keyword: "prd", "requirements" | `/shaktra:pm prd` | PRD only |
| `brainstorm` | Keyword: "brainstorm", "ideate" | `/shaktra:pm brainstorm` | Brainstorm only |
| `personas` | Keyword: "personas", "users" | `/shaktra:pm personas` | Personas only |
| `journey` | Keyword: "journey", "journey map" | `/shaktra:pm journey` | Journey only |
| `research` | Keyword: "research", "interviews", "analyze" | `/shaktra:pm research ./data/` | Research only |
| `prioritize` | Keyword: "prioritize", "rice", "rank" | `/shaktra:pm prioritize` | Prioritization only |
**Default behavior:** If input doesn't match a specific keyword, treat it as `orchestrated` with the input as context.
```
/shaktra:pm I want to build a CLI for managing dotfiles
→ Intent: orchestrated (no keyword detected)
→ Context: "I want to build a CLI for managing dotfiles"
→ Starts full PM workflow
/shaktra:pm ./docs/product-idea.md
→ Intent: orchestrated (file path, no keyword)
→ Reads document as context
→ Starts full PM workflow
/shaktra:pm prd
→ Intent: prd (keyword detected)
→ Runs standalone PRD creation
```
## Execution Flow
### 1. Read Project Context
Before dispatching any workflow:
- Read `.shaktra/settings.yml` — if missing, inform user to run `/shaktra:init` and stop
- Read `.shaktra/memory/principles.yml` — for project principles (if exists)
- Read `.shaktra/memory/anti-patterns.yml` — for failure patterns (if exists)
### 2. Classify Intent
Match user input against the intent table. Consider both explicit keywords and contextual clues.
### 3. Collect Context
**If user provided input with the command:**
| Format | Example | Action |
|---|---|---|
| Inline description | `/shaktra:pm I want to build a task queue` | Use as context, ask about research |
| Document path | `/shaktra:pm ./docs/idea.md` | Read file, ask if contains research |
| Keyword + input | `/shaktra:pm research ./interviews/` | Route to standalone workflow |
| Keyword only | `/shaktra:pm prd` | Route to standalone, ask for context there |
**If no input provided (`/shaktra:pm` alone):**
Use AskUserQuestion to guide the user. See **Guided Entry Flow** below.
---
## Guided Entry Flow
When user enters just `/shaktra:pm` with no input, use AskUserQuestion to guide them. See `guided-entry.md` for the full flow.
**Summary:**
1. **Q1: Starting Point** — "How would you like to start?"
- Describe my product idea (Recommended)
- Use my notes document
- Start from research data
- Do something specific
- _"Other" for custom input or "help me figure out"_
2. **Q2: Research Check** — "Do you have user research?"
- Yes → research-first path
- No → hypothesis-first path
3. **Guided Discovery** — For confused users, ask about their situation and route appropriately.
---
### 4. Execute Workflow
Route to the matching workflow:
| Intent | Workflow File |
|---|---|
| `orchestrated` | `full-workflow.md` (triage → `full-workflow-research.md` or `full-workflow-hypothesis.md`) |
| `prd` | `prd-workflow.md` |
| `brainstorm` | `brainstorm-workflow.md` |
| `personas` | `persona-workflow.md` |
| `journey` | `journey-workflow.md` |
| `research` | `research-workflow.md` |
| `prioritize` | `prioritization-workflow.md` |
### 5. Present Completion Report
After any workflow completes, present a structured summary (see Completion Report below).
---
## Agent Dispatch Reference
| Agent | Model | Skills Loaded | Spawned For |
|---|---|---|---|
| shaktra-product-manager | sonnet | shaktra-reference | All PM workflows |
| shaktra-memory-curator | sonnet | shaktra-reference | End of every workflow |
---
## Sub-Agent Prompt Templates
See `agent-prompts.md` for all Task() prompt templates used when spawning the PM agent and memory curator in different modes.
---
## Workflow Prerequisites
### Orchestrated Flow (default)
The orchestrated flow creates artifacts in this order — PRD is created LAST:
```
Research-First: Research → Personas → Journeys → PRD
Hypothesis-First: Brainstorm → Personas → Journeys → PRD
```
This ensures personas and journey insights inform the PRD requirements.
### Standalone Workflows
For standalone intents (when using keywords like `/shaktra:pm prd`):
| Workflow | Requires | If Missing |
|---|---|---|
| Brainstorm | `.shaktra/settings.yml` | Run `/shaktra:init` |
| Research | Research inputs (user provides) | User must provide input |
| Personas | `.shaktra/settings.yml`, context (brainstorm or research) | Will ask for context |
| Journey | `.shaktra/personas/*.yml` | Run personas first |
| PRD | `.shaktra/settings.yml`, context (brainstorm, personas, journeys recommended) | Will ask for context |
| Prioritize | `.shaktra/stories/*.yml` | Run `/shaktra:tpm` first |
---
## Completion Report
After every workflow, present:
```
## PM Workflow Complete
**Workflow:** {workflow name}
**Intent:** {classified intent}
### Artifacts Created
- {list of files created or modified with paths}
### Quality Results
- PRD review: {PASS/BLOCKED — or N/A}
- Review iterations: {count — or N/A}
### Research Summary (if applicable)
- Interviews analyzed: {count}
- Themes identified: {count}
- Recommendations: {count}
### Personas (if applicable)
- Created: {count}
- Evidence coverage: {avg evidence per persona}
### Journeys (if applicable)
- Created: {count}
- Opportunities identified: {count}
### Prioritization (if applicable)
- Quick Wins: {count}
- Big Bets: {count}
- Total stories scored: {count}
### Next Step
- {recommended next action}
```
## Memory Capture (brainstorm and research modes only)
After `RESEARCH_SYNTHESIZED` or brainstorm completion:
1. Create `.shaktra/observations/<mode>-<date>.yml` (create directory if needed)
2. Write observations:
- `type: discovery` — surprising market findings, domain constraints, user behavior patterns
- `type: observation` — requirements patterns, persona insights that inform future stories
3. Spawn memory-curator:
```
You are the shaktra-memory-curator agent. Consolidate PM workflow observations.
Observations path: {observations_path}
Workflow type: pm
Settings: {settings_path}
```
## Guard Tokens
This workflow emits and responds to:
- `PRD_COMPLETE` — PRD written to disk
- `PRD_APPROVED` — user approved the PRD (required for full workflow to continue)
- `RESEARCH_SYNTHESIZED` — research analysis complete
- `PERSONAS_COMPLETE` — all personas created
- `QUALITY_PASS` / `QUALITY_BLOCKED` — quality review results
- `CLARIFICATION_NEEDED` — agent needs user input
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |