AB promo-stock
During peak sales (Black Friday, Cyber Monday, Singles' Day, flash sales), align hot-SKU remaining inventory with live ad spend—surface sell-through risk, auto-style Pre-order messaging when stock crosses a threshold, and produce supplier restock chase communications. Use this skill whenever the user mentions inventory under ~20% while ads still run, the last 24 hours before a major promo, oversell risk, cutting Meta/Google/TikTok when stock is thin, switching PDP to pre-order, or supplier expedite emails—even if they only say "we might stock out during the sale" or "ads are still on but warehouse is low." Also trigger on safety stock breaches, SKU-level pacing, and multi-channel inventory sync. Do NOT use for annual financial inventory audits with no live promo or ads context, pure demand forecasting with no immediate copy or comms need, or legal product compliance for pre-orders in regulated categories without the user asking for operational templates.
As a process B 65/100 · Nearly there — weak spots: result and completion, inputs and preconditions, consistency
How to improve
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 6. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 65/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 3 mutating operations with no state check
- 40Consistency. Frontmatter name (promo-stock) differs from the folder (promp-stock)
- 100Tools and files. No external tools needed
- 100Steps. 12 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 3 branches, has a failure section
- 100Execution cost. Instruction body is 764 tokens
- medium 3 test cases, all positive: not one "should refuse" or "should ask first"
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
- +3Description length 968: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +4No input/output examples
- +1No license
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 9 headings
- +3Step-by-step instructions: 12 items
- +4Reference files are cited in the instructions (1 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 90.