بنقرة واحدة
pm-craft
Router for the pm-craft plugin — matches your need to the right skill and tells you what to type.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Router for the pm-craft plugin — matches your need to the right skill and tells you what to type.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Builds a self-contained interactive HTML visualisation that makes a process, algorithm, decision logic, or system experientially understandable.
Write a per-topic handoff file and log it in a single index at the end of a working session, so any topic can be resumed later in a fresh session. Use when the user wants to wrap up or park a session ("handoff", "wrap up", "park this"), or to resume or pick up a previous topic. Also suggest it proactively when a substantive session is clearly winding down.
Simulates a Visionary-vs-Skeptic debate over a product idea, then synthesises a path forward.
Use the Obsidian CLI to interact with the vault natively—search, read, create, manage tasks, properties, and daily notes. Prefer CLI over raw file I/O for vault-aware operations.
Use when stress-testing a product idea before committing — interviews the user on six PM lenses and produces a conviction brief alongside the session.
Analyses a problem through mental-model lenses — 11 personas and 20 frameworks, four modes (Tuning, Solo, Counterpoint, Harmony), Simple and Expert tracks.
| name | pm-craft |
| description | Router for the pm-craft plugin — matches your need to the right skill and tells you what to type. |
| disable-model-invocation | true |
Match the user's need to one skill below and tell them what to type. Most of these are user-invoked: you cannot fire them yourself — recommend the command, don't attempt the invocation. If the user passed their problem as an argument, recommend one skill with a sentence of rationale (two skills only if genuinely torn), then stop.
| The user wants to… | Skill | How it fires |
|---|---|---|
| Stress-test a product idea by being interviewed — six PM lenses, produces a conviction brief | product-grill | Fires on its own when you describe an idea to stress-test |
| Widen the lens on a strategic problem — Cynefin, Three Horizons, Wardley, Idealised Design | thinking-partner | Fires on its own ("think bigger", "sanity check", or name a framework) |
| Watch an idea attacked and defended — simulated Visionary-vs-Skeptic debate, then a synthesis | product-debate | Type /product-debate |
| Build a Lean Product Canvas, coached Socratically section by section | lpc-author | Type /lpc-author |
| A narrative, step-by-step walkthrough document of code, docs, or a system | linear-walkthrough | Type /linear-walkthrough |
| An interactive HTML visualisation of a process, algorithm, or decision logic | interactive-explainer | Type /interactive-explainer (best fed a walkthrough) |
| Analyse a problem through mental-model lenses — 11 personas, 20 frameworks, four modes | mind-ensemble | Type /mind-ensemble |
| Wrap up a session into a resumable per-topic handoff file, or resume a parked topic | session-handoff | Fires on its own ("wrap up", "park this", resume requests) |
Disambiguation for the three that overlap: product-grill interviews you (your conviction is the subject); product-debate argues at you (the idea is the subject, you watch); mind-ensemble analyses through chosen lenses (the lens does the work, alone or composed).