BB az-eu-website-privacy-audit
Audits a website for compliance with Azerbaijan's Law on Personal Data No. 998-IIIQ and, where applicable, EU GDPR plus ePrivacy/cookie consent rules. Inventories the privacy documents present (privacy policy, cookie policy, cookie banner, consent flow, controller and DPO contact, data subject rights channel, cross-border transfer disclosures, AZ operator-registration references) and scores each against the applicable statutory requirements. Produces a dual-layer report: a plain-language traffic-light summary for business owners plus a clause-by-clause findings table with article-level citations for lawyers. Assessment-only — no drafting. Use whenever the user shares a URL or privacy/cookie policy text for an AZ-based or AZ-targeted site; also when the user mentions an .az domain, Law 998, the AZ State Register, ePrivacy, an Art. 27 EU representative, or asks "is my site GDPR compliant", "do I need to register as an operator in Azerbaijan", or "is our cookie banner lawful" — even without the word "audit".
Audits a website for compliance with Azerbaijan's Law on Personal Data No.
As a process B 71/100 · Nearly there — weak spots: consistency, running it twice
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
- 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 · 2
✓ No critical or high findings
Medium and low: 2
-
medium Exfiltration
intent-browser-credential-storereferences/gdpr_for_az_websites.md:41Accesses a browser credential / cookie store (documentation of a security skill)- Cookies set to support advertising networks operating in the EU.
security skill
A further 1 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 2, column 14: description: Audits a website for compliance with Azerbaijan's Law on Personal … ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - note
edit-residuethe text marks something as outdated (lines 112): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 71/100
- 30Running it twice. 1 mutating operations with no state check
- 40Consistency. Frontmatter name (az-eu-website-privacy-audit) differs from the folder (az-eu-website-privacy-audit-mirza-chiragov)
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 70Failures and branches. 8 branches
- 85Steps. 59 steps, 1 vague phrases
- 100Inputs and preconditions. Inputs and preconditions are listed
- 100Execution cost. Instruction body is 3525 tokens
- 100Progress reporting. Reports progress
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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1020: 120–800 characters recommended
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +4Structure: 17 headings
- +3Step-by-step instructions: 59 items
- +3Output format is stated explicitly
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.