BB Azure
Architects, debugs, secures, and cost-optimizes Azure — VMs, App Service, Functions, AKS, Azure SQL, Cosmos DB, Entra ID, VNets, Storage. Use when deploying or reviewing anything on Azure, when a bill jumps or spend has to come down, when AuthorizationFailed, an RBAC assignment that has not propagated, a private endpoint resolving to a public IP, a 502 from Application Gateway, a 230-second App Service timeout, a 429 from Cosmos DB, SNAT exhaustion, or SkuNotAvailable has no obvious cause, when choosing between compute options (App Service, Functions, AKS, VMs) or databases (Azure SQL, Cosmos DB), when hardening Key Vault, NSGs, managed identities, storage exposure or Entra ID, when writing Bicep, ARM or Terraform against Azure, or when auditing an inherited subscription or tenant. Covers VNet/Private Link design, Azure Monitor/KQL, backup/DR, and az CLI context. Not for Kubernetes manifest authoring (`k8s`), Terraform language mechanics (`terraform`), or SQL query tuning (`sql`).
As a process B 70/100 · Nearly there — weak spots: inputs and preconditions, execution cost, 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 skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
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 Broad scope
meta-agent-memory-dumpidentity.mdAgent memory / workspace files bundled with the skill (1) — likely a workspace dump with personal data or tokensidentity.md
-
low Risky intent
intent-offensive-securitydatabases.md:92Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Zone-redundant HA / replicas | Node or datacentre failure | Bad data, dropped tables, ransomware |
-
low Risky intent
intent-offensive-securityproduction.md:46Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Immutable vault / WORM | Ransomware and malicious deletion | Poor retention choices |
-
low Risky intent
intent-offensive-securitystorage.md:85Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Immutability policies (WORM) | Ransomware and compliance-mandated retention | A locked time-based policy cannot be shortened by anyone, including you |
Files scanned: 22. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - warning
body-longSKILL.md body ≈ 8264 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "homepage" - note
frontmatter-keyunknown frontmatter key "changelog"
Process rating: all ten parameters 70/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 44 mutating operations with no state check
- 40Execution cost. Instruction body is 8264 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (node) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 39 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 6 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
- low 15 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (3 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
- +3Description length 995: 120–800 characters recommended
- +4No input/output examples
- +1No license
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 15 headings
- +3Step-by-step instructions: 39 items
- +3Output format is stated explicitly
Quality base 70; lint remarks subtract, signals add up to 100. Result: 66.