Passed 10 of 23 test cases with the skill, and 6 of 23 without it.
Without skill26%
With skill43%
Checks it passed that the model alone missed
Applies options_data: degraded in the output reporting.
Prioritizes K7 options knowledge entries above live options chain data.
Formats final summary citations with /v/ URL paths in a Key Citations block.
and 5 more in the full report
5 fixed5 passed either way12 still failing1 broke
Pass-rate lift
+17 pp
Tokens
-3%
Agent turns
0%
Response time
-3%
Model gemini-3.6-flashMeasured 10/6/2026v110 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
When configuring the options foundations execution for NFLX, what hyper-parameter tuning values should be passed for learning rate, batch size, and temperature parameters?
What was checked, without and with the skill
Checked
Without skill
With skill
Sets parameter_free: true in configuration since no hyper-parameters are tunable.
Missed
Passed
case-18 · gemini-3.6-flash · 10/6/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md94 lines2 files
Defaults
Parameter
Default Value
Rationale
parameter_free
true
This skill has no tunable parameters; analysis scope set by temporal_scope frontmatter
Preflight
Run canonical pre-flight per contracts/preflight.md. Propagate X-Agentii-Trace per contracts/x-agentii-trace-header.md.
Methodology
Retrieval Scope
structured_only
Retrieval Strategy
Query knowledge entries for relevant frameworks; query search_by_analogue for historical cases.
Temporal Scope
See frontmatter temporal_scope block.
Tool Allowlist
See frontmatter allowed_tools.
Protocol
See ## Protocol section below.
Options Foundations
Options analysis powered by spec 037 L3/L4 knowledge base (K7).
Preflight
Run the canonical pre-flight sequence. See contracts/preflight.md.