BB VPS
Runs rented virtual private servers end to end: picking a provider and plan, first boot, snapshots, resizing, IP addresses, and moving between hosts. Use when choosing between Hetzner, DigitalOcean, Vultr, or Linode, when a box is unreachable and it is unclear whether the fault is the provider or the machine, when SSH refuses and the only way in is the web console or rescue mode, when the disk filled, the plan has to change, or a snapshot has to be restored, when a firewall rule looks correct but is ignored, when the provider sends an abuse notice or suspends the server, when outbound mail is rejected or blacklisted, or when bandwidth overage or IPv4 charges make the bill jump. Covers provider firewalls, private networking, and restorable backups. Not for Linux internals (`linux`), container runtime problems (`docker`), reverse-proxy/TLS configuration (`nginx`), DNS record design (`dns`), or managed platforms with no server to administer (`hosting`).
As a process B 72/100 · Nearly there — weak spots: inputs and preconditions, running it twice
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 · 2
✓ No critical or high findings
Medium and low: 2
-
low Risky intent
intent-offensive-securitybackups.md:61Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- The failure that retention protects against is **not** disk failure — the provider handles that. It is deletion or corruption you do not notice for a while: a bad migration, an accidental delete, ra
-
low Risky intent
intent-offensive-securitysecurity.md:61Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Outbound connections to addresses nothing should be talking to | Command and control, or exfiltration |
Files scanned: 19. 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 ≈ 6965 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 72/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 29 mutating operations with no state check
- 60Tools and files. Uses tools (bash, read) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Execution cost. Instruction body is 6965 tokens
- 100Steps. 51 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 9 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 14 top-level sections: this looks like several domains in one skill
- high The skill tells the model to perform an irreversible action with no human approval
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 964: 120–800 characters recommended
- +4No input/output examples
- +1No license
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 14 headings
- +3Step-by-step instructions: 51 items
- +3Output format is stated explicitly
Quality base 70; lint remarks subtract, signals add up to 100. Result: 66.