AB hudl-model-switch
Switch between LLM models on the Huddle01 GRU gateway. Use this skill whenever the user mentions switching models, changing models, upgrading, downgrading, "switch to opus", "use minimax", "use claude", "use the smart model", "use the cheap model", "change model", "swap model", or any variation of wanting a different LLM. Also trigger when the user asks "what model am I on", "which model", or "current model". This skill ONLY works with the hudl provider backed by gru.huddle01.io.
As a process B 79/100 · Nearly there — weak spots: result and completion
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 · 1
✓ No critical or high findings
Medium and low: 1
-
low Exfiltration
net-credential-usereferences/models.md:101Credential used in a network call (verify the destination is the intended service) (the skill's own vendor host; quoted — discussed, not commanded)3. Verify available models via: `curl -s https://gru.….io/v1/models -H "Authorization: Bearer $KEY" | jq '.data[].id'`
vendor-hostquoted
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 79/100
- 0Result and completion. Does not say what the result is
- 70When it triggers. States when to use, but not when not to
- 70Failures and branches. 7 branches
- 100Tools and files. No external tools needed
- 100Steps. 54 steps
- 100Inputs and preconditions. Inputs and preconditions are listed
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1754 tokens
- 100Running it twice. No mutating operations
- 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 (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
- +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
- +1No license
- +2Single-language instructions
- +5Description quotes 10 example trigger phrases
- +3Description length 484: enough signal without eating the budget
- +4Structure: 7 headings
- +3Step-by-step instructions: 54 items
- +4Has examples (7 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 92.