docs(skills): refine snapshot diff and QuickJS guidance - #350

Merged

rishabhraj36 merged 28 commits into

mainfrom

Aug 18, 2026

Commits on Aug 10, 2026

benchmakrs added
rishabhraj36
committed
938a791
fix: removed missing refs from readme
rishabhraj36
committed
b7e4d5e
feat(benchmarks): support OpenAI as judge provider
rishabhraj36
committed
bfa4d4b
fix: screenshot upload
rishabhraj36
committed
2c5c778

Commits on Aug 13, 2026

Merge branch 'main' into feat/evals
rishabhraj36
committed
c861319
feat: added cost support
rishabhraj36
committed
134c8a0
docs: design benchmark-specific judge prompts
beubax
committed
712bc99
feat(benchmarks): split general and stealth judge prompts
beubax
committed
332db0f
Delete docs/superpowers/specs directory
beubax
authored
57fb122
chore: bump version metadata for 0.7.0 release
beubax
committed
c69011f
Revert "chore: bump version metadata for 0.7.0 release"
beubax
committed
ac8d5fa

Commits on Aug 14, 2026

feat(benchmarks): expose full Webcmd skill pack to Codex evals
rishabhraj36
committed
dce482c
Merge remote-tracking branch 'origin/main' into feat/evals
beubax
committed
4234b94
feat(benchmarks): clarify stealth captcha handling
beubax
committed
d7549e7
Merge remote-tracking branch 'origin/feat/evals' into feat/evals
beubax
committed
0cb090c
fix: added support for webcmd browser skills only
rishabhraj36
committed
317e66c
feat(benchmarks): run eval attempts in parallel
beubax
committed
268d8b8
fix(benchmarks): isolate parallel webcmd sessions
beubax
committed
9af8437
test(benchmarks): cover parallel webcmd sessions
beubax
committed
251f4ef
pi support (#313)
rishabhraj36
authored
3f50d78
chore(benchmarks): pin webcmd 0.7.1
beubax
committed
6cdf4f2
refactor: disable concurrency in benchmark evaluation and remove integration tests
beubax
committed
59581e0
feat(benchmarks): support Codex subscription evals
beubax
committed
8d99e67

Commits on Aug 18, 2026