quit-sponsor

What it does

Turns an AI agent with persistent memory into a quit-smoking sponsor. It operates as a witness with long memory, on call at moments a human sponsor cannot be. The sponsor orchestrates evidence-based tools (immediate execution of the quit decision, urge surfing, slip attribution coaching, withdrawal timelines, nutrition and alcohol rules, NRT guidance, a two-year aftercare cadence) plus a sponsor decision tree with an order-of-operations for colliding rules, a three-clause contract, purge ritual, graded live-crisis ladder for a pack already in hand, wave protocol, slip and post-relapse protocols, recurrence escalation, silence protocol, a Ulysses pact, a red-flag medical playbook, high-risk situation mapping, if-then plans, and a timestamped logbook. Includes a low-verbal client mode and an optional module for cannabis co-use and tobacco-mixed joints. Not a medical device.

How it works

The skill emphasizes activation and role: offer the sponsor role when the person is quitting or asks for help, and only take it if they accept or request it. It requires persistent memory across sessions to maintain the logbook, contract, and risk map. It insists on memory-backed receipts for all events (slips, pride, etc.) and uses a three-clause contract and purge rituals as part of immediate-quit procedures. It specifies running on strong models and includes references to a crisis-testing framework (MODEL_FIT.md) for crisis handling. The science section operationalizes rules: do not schedule quit dates; when the person announces a quit decision with real charge, execute within the hour with purge ritual and a T-0 timestamp and open the logbook; pre-quit training to prepare for a click; if not ready, build toolkit in layer 2. It outlines withdrawal timelines (0-72 hours, weeks 1-4) and related behaviors (caffeine adjustment, hydration, taste changes), and provides rules for hunger, water, alcohol, weight, movement, and nicotine replacement. It also details how to handle high-risk situations and escalation paths. It describes the per-event guidance for cravings and the step-by-step responses such as offering a brisk movement as a first response to cravings, and how to frame NRT and pharmacotherapy options as clinician territory. It notes that the vaping stance varies by law and requests checking local regulation.

When to use it

Activate when a person asks for help quitting, announces they are quitting, reports a craving, a slip, or a relapse, goes silent mid-quit, or asks the agent to witness and track a quit. The activation triggers the sponsor role and begins logging and protocol execution.

What it can touch

Tools: claude-code. The skill requires persistent memory to maintain a logbook, contract, and risk map across sessions.

Caveats

Not a medical device. Safety considerations reference SAFETY.md; the rule set includes medical red flags and escalation to medical help when red flags appear (fever, chest pain, etc.). The text notes clinician territory for pharmacotherapy and that professional support with medication remains strongest evidence; the sponsor is a continuity layer, not a replacement. The module mentions an optional cannabis co-use module and tobacco-mixed joints, and cites live field data and simulations for certain post-relapse states as forecasted states rather than guaranteed outcomes.