BC autoecom
Turn any ecommerce store into daily Instagram & TikTok product carousels. AI pipeline: agent auto-extracts brand kit (logo, colors, font, voice) from the store URL via WebFetch + vision, picks the next bestseller round-robin, generates 3–8 stylized slide images with nano-banana (Gemini 2.5 Flash Image) using the real product photo as reference, composes branded slides with Pillow, you approve, Upload-Post publishes. Use when the user wants daily product carousels, mentions autoecom, ecommerce content, product slides, shop content automation, or asks for the daily carousel batch.
As a process C 62/100 · Has gaps — weak spots: result and completion, inputs and preconditions
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
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 · 4
✓ No critical or high findings
Medium and low: 4
-
medium Exfiltration
exfil-send-secrets-to-urlSKILL.md:17Instruction to send secrets/history to an external endpoint (URL is a reference link, not a destination; the skill's own vendor host; quoted — discussed, not commanded)- { name: UPLOAD_POST_API_KEY, required: true, description: "Upload-Post API key (https://app.upload-post.com → Settings → API Keys)" }reference linkvendor-hostquoted -
low Exfiltration
read-dotenvREADME.md:42Reads a .env fileSet up https://github.com/mutonby/skill-autoecom for me. Read README.md and SKILL.md, clone the repo into ~/Documents/skill-autoecom, create the venv, install requirements.txt, copy .env.example to .e
-
low Exfiltration
read-dotenvREADME.md:241Reads a .env filecp .env.example .env
-
low Exfiltration
net-credential-useSKILL.md:64Credential used in a network call (verify the destination is the intended service) (the skill's own vendor host; quoted — discussed, not commanded)- Verify: `curl -H "Authorization: Apikey $UPLOAD_POST_API_KEY" https://api.upload-post.com/api/uploadposts/me`.
vendor-hostquoted
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 6734 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 6734 tokens
- 100Steps. 80 steps
- 100Failures and branches. 17 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (15 tags): a typed call is more reliable
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)
- +3Output format is not stated: the model decides each time
- -4Absolute local paths (C:\Users, /home/…): not portable
- +2Single-language instructions
- +3Description length 585: enough signal without eating the budget
- +4Structure: 29 headings
- +3Step-by-step instructions: 80 items
- +4Has examples (13 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 71.