Skip to content

feat(sn60): add bohdansolovie-20260723-09 (beat Dexterity104 #197)#199

Closed
bohdansolovie wants to merge 1 commit into
Autovara:mainfrom
bohdansolovie:submission/bohdansolovie-20260723-09
Closed

feat(sn60): add bohdansolovie-20260723-09 (beat Dexterity104 #197)#199
bohdansolovie wants to merge 1 commit into
Autovara:mainfrom
bohdansolovie:submission/bohdansolovie-20260723-09

Conversation

@bohdansolovie

Copy link
Copy Markdown
Contributor

Summary

  • Defends the crown against #197 (Dexterity104-20260723-01), which already beat a prior king on project-pass consistency.
  • Takes feat(sn60): add Dexterity104-20260723-01 #197's decisive levers — temperature 0, independent re-audit of the same central files, confidence floor 0.45, emit 14 / 4-per-file — and keeps our cross-pass vote boost + localization probes.
  • Uses the full ~840s agent wall (deadline 780s) so map → deep → reaudit reliably finish on every replica; wide pass only if clock remains.
  • Fresh Chutes seal bound to this exact bundle (no key reuse).

Test plan

Made with Cursor

…lovie-20260723-09)

Absorb Dexterity104's consistency levers (temp=0, independent same-file
re-audit, conf floor, tight emit) and keep cross-pass vote ranking plus
structural probes, sealed fresh to this bundle.

Co-authored-by: Cursor <cursoragent@cursor.com>
@carlos4s carlos4s added kata:pending Kata screened this PR: waiting for the next competition round. kata:executing Kata is scoring this challenger in the current round. kata:losing Kata evaluated this challenger and it did not beat the king. and removed kata:pending Kata screened this PR: waiting for the next competition round. kata:executing Kata is scoring this challenger in the current round. labels Jul 23, 2026
@carlos4s

Copy link
Copy Markdown
Member

Thanks for the challenge! Your agent went up against the current king and came up just short this time, so I've closed this PR. Here's exactly how it played out:

Here's how you stacked up against the king's average over its last 3 challenges, in the order the crown is decided — reading top to bottom, you take the crown by clearly beating the king on a line — by more than that line's margin — without falling behind on a higher one:

# Signal You King Win margin
1 Project pass score
projects solved on a 2-of-3 majority
0% (0/7) 5% >14.3% ← decided here
2 Projects passed
projects with at least one passing run
0 0.3 >1
3 True positives
real bugs found out of bugs planted
3/37 2.3 >2
4 Fewer invalid runs
runs that errored out — fewer is better
0 3.7 >2
5 Precision
how many of your findings were real
100.0% 55.6% >5%
6 F1 score
overall balance of precision and detection
15.0% 13.6% >5%

How the crown is decided: read the lines top to bottom. On each line — if you're behind the king, the king keeps the crown and you can't make it up on a lower line; if you clearly beat the king (by more than that line's margin) you take the crown; if you're close (within the margin), that line is a tie and it moves to the next one. So to take the throne you have to clearly beat the king on some line (pass score first) without falling behind on a higher one. This stops a single lucky run from flipping the crown.

(The king's numbers are averaged over its whole reign — 3 challenges — since that's what it's judged on, so beating it on a single run isn't enough.)

So it came down to project pass score (tier 1) — that's where you fell behind the king, and since you can't make up a deficit on a lower line, the king kept the crown. To take it you'd need to stay even with the king (within the margin) down to a line where you clearly beat it.

Tweak your agent and open a fresh PR whenever you're ready — you can absolutely take another run at the throne. 👑

@carlos4s carlos4s closed this Jul 23, 2026
@carlos4s carlos4s added the kata:lane:sn60__bitsec Kata competition lane this PR belongs to (membership only, not a routing input). label Jul 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

kata:lane:sn60__bitsec Kata competition lane this PR belongs to (membership only, not a routing input). kata:losing Kata evaluated this challenger and it did not beat the king.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants