ibm777p2

Intelligence Emotions — your mental-fitness team. Five coaches, one practice: catch the saboteurs, do the reps, let the Sage drive. This root skill is the front door: it meets whatever the user brings and routes to the right session. Use when the user invokes "pq", asks "where do I start", wants to know what the team can do — or brings something that doesn't obviously fit one skill.

Closes a growth-spec at the end of its horizon. Reviews what was actually practiced across the 21 days, writes the closing journal entry, marks the commitment committed, and — if the user wants — seeds the next one. Use when asked to "close out my spec", "commit", "the 21 days are up", or when /habit-watch or /pq-retro notices an open commitment past its horizon.

The daily routine, chained: morning intention → saboteur forecast for the day ahead → (evening) hand-off to /pq-retro. Morning mode sets one intention and pre-names the day's likely ambushes; evening mode runs the retro. Use for "good morning", "start my day", "daily", "morning check-in" — or in the evening, "end my day", which routes straight to /pq-retro.

Turns a vague intention ("be more patient with my kids") into a structured 21-day practice: the intention in the user's words, a trigger inventory, daily reps anchored to real moments, and measurable check-ins — saved as an open commitment in the journal. Use when the user states an intention without a plan: "I want to be more...", "I need to stop...", "growth spec", "help me actually do this". When the 21 days end, /commit closes it out.

The post-commitment check-in loop. After a new practice starts, this is the scheduled look-in: reads the journal since the last check, watches for early drift signals (anchors going quiet, catches turning hindsight-only), and re-fits the practice while adjustments are still cheap. Use for "check in on my habit", "habit watch", "how's my practice going", or on the cadence the user configured. Watches an open practice mid-flight; /commit closes it at the horizon.

Writes up an insight or breakthrough as a keeper document — either a practice guide (how to run this move again when it's needed) or an explanation (why this pattern formed and what loosens it). Insights decay; documents don't. Use when something landed: "write this down", "I want to keep this", "insight doc", "that just clicked" — usually right after a session where a realization happened.

Root-cause debugging of a negative thought or feeling. Systematically traces trigger → saboteur → the lie it told → the underlying need, then prescribes a contextual PQ rep and logs the interception to the journal. Use when the user brings one specific bad moment: "I feel worthless", "I can't stop replaying that meeting", "why did that comment wreck my day", "intercept this". The right tool when something specific stung — for a whole life challenge, route to /sage-session instead.

From-scratch design of a personal operating system: the morning routine, rep triggers stacked onto existing anchors, and environment design — so the practice runs on rails instead of willpower. Produces a one-page design the user owns. Use when asked to "design my routine", "set up my system", "life design", "build my morning" — or after a few weeks of practice, when the pieces deserve an architecture.

VentureStack — your AI founding team, before you have one. Six mentors, twelve disciplines, one practice: build the judgment and habits of an entrepreneur. This root skill is the front door: it greets, runs first-session onboarding, detects what the user needs, and routes to the right mentor. Use when the user invokes "vs", asks "where do I start", wants to know what the team can do — or brings something that doesn't obviously fit one skill.

Learn the present AI landscape through the canon: a dated snapshot of the model ecosystem plus worked case studies that map current events onto the twelve disciplines — one case study per session, connected to the user's own venture. Use when asked "what does AI mean for my business", "AI brief", "should I worry about these model releases", "catch me up on the AI landscape".

Design one small bet: a cheap, low-risk, quickly-judged experiment with a cost cap, a time cap, and explicit kill/scale criteria — so the user finds out what actually works before committing big resources. Also runs the resolution session when a bet's time cap expires. Use when asked "test this idea", "design a bet", "small bet", "should I quit my job for this" (the answer starts with a bet), or when /vs-sweetspot hands over a hypothesis.

The flagship. Build the user's Compass: core values (what they'd hold even if it cost them), core purpose (why the venture exists beyond money), and a Summit Goal — one clear, huge 10-to-25-year objective plus a vivid written picture of what reaching it looks and feels like. Runs across one or more sessions; partial progress is saved and resumed. Use when asked "why am I doing this", "what's my vision", "compass session", "long-term direction" — or when another session keeps hitting the question underneath the question.

Map the user's Momentum Engine: the 4-to-6-step self-reinforcing loop where each step almost inevitably drives the next. Name each step, test every link, find where the energy leaks, and commit to turning the loop with consistency instead of lurching between new programs. Use when asked "nothing compounds", "map my engine", "momentum engine", "why does every month feel like starting over".

The brutal-facts session: confront what is true right now that the user isn't acting on — without flinching and without despair, holding unwavering resolve alongside unflinched reality. Use when asked "give it to me straight", "reality check", "am I kidding myself", "brutal facts" — or when optimism and the numbers have visibly parted ways.

Search and browse the private journal: decisions and their reasoning, bets and their outcomes, the march record, the luck log. Also the recovery room — "what did I say last month?", "why did I decide that?", "show me every bet I've killed". Handles supersede and redact requests with absolute respect. Use when the user wants to look back, find a past record, correct one, or be forgotten about something.

Log a luck event — good or bad — and grade the response, never the draw. You can't control what happens to you; you control what you do with it inside the window while it's still alive. Use when something just happened TO the user: a competitor stumbled, a platform changed the rules, a chance meeting, a supplier folded, a windfall, a setback nobody caused.