End-of-work audit that identifies which governance artifacts should be updated based on actual changes.
SkillScore77Maintained
Automatic
◆Doc Ops pack
Last verified 8/3/2026
What it changes
Passed 17 of 22 test cases with the skill, and 6 of 22 without it.
Without skill27%
With skill77%
Checks it passed that the model alone missed
Required updates explicitly lists ARCHITECTURE.md
The Trigger summary includes '- Change type: docs'
The Trigger summary includes '- Change type: bugfix'
and 12 more in the full report
11 fixed6 passed either way5 still failing0 broke
Pass-rate lift
+50 pp
Tokens
-31%
Agent turns
0%
Response time
-45%
Model gemini-3.6-flashMeasured 8/3/2026v117 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 11
Prompt
Fixed three typos in the install instructions of `README.md`. No code or structural changes were made. Generate a doc alignment assessment report. I think we should mark this as requiring a decision.
What was checked, without and with the skill
Checked
Without skill
With skill
The Trigger summary includes '- Change type: docs'
Missed
Passed
The Trigger summary includes '- Decision required: no'
Missed
Passed
The Trigger summary includes '- Structural change: no'
Missed
Passed
case-08 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md45 linessingle file
Governance Sync
Goal
Keep governance artifacts truthful with minimal churn.
Non-negotiables
Read-only. Do not edit files automatically.
Recommend updates only when triggered by actual changes.
Be explicit about what changed and why that forces a doc update.
Workflow
Inspect what changed
Prefer git diff/summary when available, otherwise use the user’s change summary.
Determine which artifacts’ truth conditions changed
Common candidates:
Access Kalshi prediction market data including market prices, orderbooks, trades, events, and series information. Use when the user asks about prediction markets, Kalshi markets, betting odds, market prices, or needs to search or analyze prediction market data.