Passed 11 of 19 test cases with the skill, and 4 of 19 without it.
Without skill20%
With skill60%
Checks it passed that the model alone missed
Directs the user to agentii.ai/api-keys
Requires at least 1 citation per 200 words
Updates the index file at sessions/INDEX.md
and 11 more in the full report
13 fixed2 passed either way7 still failing3 broke
Pass-rate lift
+40 pp
Tokens
+22%
Agent turns
0%
Response time
-57%
Model gemini-3.6-flashMeasured 10/6/2026v115 of 25 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 13
Prompt
In the mandatory 5-part report structure, what is the exact title and relative position of the second section?
What was checked, without and with the skill
Checked
Without skill
With skill
Titles section 2 as Data Sources and places it immediately after Executive Summary
Missed
Passed
case-21 · 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.md98 lines4 files
revenue-decomp
Triggers
Revenue decomposition
segment breakdown
geographic revenue split
product-line waterfall
revenue mix analysis
business segment performance
divisional revenue
revenue concentration
customer revenue dependency
channel revenue analysis
Defaults
Parameter
Default Value
Rationale
ticker
(required)
Stock symbol to analyze
lookback_quarters
4
Standard lookback for this skill type
Methodology
1. Retrieval Scope
This skill operates with retrieval_scope: structured_only. It performs structured data retrieval only (XBRL facts, financials, earnings calendar) — no unstructured document search. Document-retrieval tools are excluded from allowed_tools.
2. Retrieval Strategy
Follows the retrieval strategy decision tree in contracts/retrieval.md. Primary branch: (a) Structured Data Query. Resolve the canonical ticker first (exact → fuzzy alias → share-class) before any data call.
3. Temporal Scope
Default lookback: 4 fiscal quarter(s); maximum: 8. The default balances recency against the trend window this analysis requires.