pj ca8b517d98 test(runner): the gate keeps looking until its budget runs out ...

Locks the three facts the campaign was missing: a window that draws after more
polls than the old count allowed still clears the gate, an app that never comes
forward ends the run with a typed error, and both the startup verdict and every
mid-run step the guard could not recover are readable off trace.jsonl.

2026-08-17 14:35:37 +05:30

..

_oracle

iOS simulator driver: Go-native companion-backed backend (#62)

2026-06-08 19:10:54 +05:30

android

correctness fixes from the first real folio dispatch, and spec authoring skills (#82)

2026-08-16 13:23:53 +05:30

bundler

llm action-selection backend (#68)

2026-07-31 21:12:00 +05:30

driver

merge origin/master into llm-recording-and-analysis

2026-08-16 18:10:55 +05:30

hierarchy

merge origin/master into llm-recording-and-analysis

2026-08-16 18:10:55 +05:30

ios

Physical-device iOS support (#64) (#66)

2026-06-09 18:38:52 +05:30

llmclient

feat(llmclient): parse usage and the served model

2026-08-12 22:56:58 +05:30

ltl

merge origin/master into llm-recording-and-analysis

2026-08-16 18:10:55 +05:30

replay

iOS simulator driver: Go-native companion-backed backend (#62)

2026-06-08 19:10:54 +05:30

runner

test(runner): the gate keeps looking until its budget runs out

2026-08-17 14:35:37 +05:30

seedspec

refactor(seedspec): move seed spec parsing out of the campaign command

2026-08-16 17:44:57 +05:30

sidecarassets

experiment instrumentation: step budgets, arm labels, campaign runner (#72)

2026-08-12 22:20:31 +05:30

testrun

merge origin/master into llm-recording-and-analysis

2026-08-16 18:10:55 +05:30

trace

feat(trace): a step can name the precondition it could not meet

2026-08-17 14:35:19 +05:30

tracecorpus

feat(tracecorpus): load recorded runs for offline measures

2026-08-16 17:44:52 +05:30

verifier

merge origin/master into llm-recording-and-analysis

2026-08-16 18:10:55 +05:30