BC quit-sponsor
Turns an AI agent with persistent memory into a quit-smoking sponsor. Use when a person asks for help quitting smoking (cigarettes or other smoked tobacco), 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. Provides evidence-based protocols (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.
Turns an AI agent with persistent memory into a quit-smoking sponsor.
As a process C 57/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 1
✓ No critical or high findings
Medium and low: 1
-
low Secrets in code
secret-high-entropy-tokenREADME.md:66High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- Ethereum: `0x72…269`
quoted
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 18191 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 57/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 10Execution cost. Instruction body is 18191 tokens: crowds the task out of the window
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 27 mutating operations with no state check
- 100Tools and files. No external tools needed
- 100Steps. 147 steps
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1009: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +4No input/output examples
- +2Single-language instructions
- +4Structure: 43 headings
- +3Step-by-step instructions: 147 items
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 68.