Validate committed feature-branch changes through the no-mistakes pipeline: intent, rebase, review, test, docs, lint, push, PR, and CI.
SkillScore86Maintained
Automatic
◆DevOps & SRE pack
Last verified 8/18/2026
What it changes
Passed 16 of 19 test cases with the skill, and 8 of 19 without it.
Without skill41%
With skill86%
Checks it passed that the model alone missed
Recommends no-mistakes axi abort
Recommends no-mistakes axi status
Recommends executing no-mistakes init
and 7 more in the full report
10 fixed9 passed either way3 still failing0 broke
Pass-rate lift
+45 pp
Tokens
-52%
Agent turns
0%
Response time
-73%
Model gemini-3.6-flashMeasured 8/18/2026v119 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 10
Prompt
We need to inspect the current state of an active local gate evaluation session without modifying any findings. What CLI subcommand displays this status?
What was checked, without and with the skill
Checked
Without skill
With skill
Recommends no-mistakes axi status
Missed
Passed
case-12 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md49 linessingle file
no-mistakes
no-mistakes is a local git gate. It validates committed branch changes before
they reach upstream and reports decision points in machine-readable TOON.
Preconditions
Work is committed.
Current branch is a feature branch, not the default branch.
The repository has been initialized with no-mistakes init.
The intent is explicit: what the user wanted to accomplish, not only a diff
summary.
Run
no-mistakes axi run --intent "<what the user set out to accomplish>"
Use the user's exact objective as the intent, enriched with material decisions
and constraints learned during the work.
Decision Loop
If output contains a gate: object, inspect each finding:
auto-fix: make the mechanical fix, then respond with fix.
no-op: approve or continue when nothing needs changing.
ask-user: stop and relay the finding to the user before responding.