Milestones
List view
What makes this runnable indefinitely: cadence and budget, durable transcripts, cost reporting, and the product findings this repository cannot close itself.
No due date•0/11 issues closedConsole has no scenarios, and five benchmark scenarios have never been failed by any agent (#47) — three of them Event Gateway, so this was never an Outpost problem; it was invisible until scenario-criteria counted. A product score drawn from tests nobody fails is not a product score. outpost-001 and outpost-003 are addressed; the rest are being sorted into harden, floor or redesign from transcripts rather than from argument.
No due date•1/4 issues closedSkills must improve the product or there is no point having them. The weak model's delta has now read -3 (13 Aug), -2 (late Aug), -1 (1 Sept) and +4 (14 Sept), the last on a clean pairwise comparison with no harness change between it and the run before. The sign is no longer replicating, so neither the magnitude nor the direction is established. Claude reads +2 and GPT-5.6 0 on the last full matrix. Settling it needs both arms at three attempts and cannot be cheaply targeted: 9 of 57 paired cells differ on any given run, but 14 of 19 scenarios have disagreed at some point, so there is no stable subset where skills matter.
No due date•3/6 issues closedThe page is live and publishes from a release rather than from whatever ran last. What remains is the release helper, and clearing the publishing hold once a snapshot exists that was measured end to end under the current base prompt. The two explanatory gaps that shipped knowingly — the paired columns and the deliberately weaker model — were closed as not planned on 29 August: the page carries the scoreboard, and anyone wanting method goes to this repository.
No due date•6/7 issues closedSubstantially done: three independent causes of moving results found and handled. What remains will not change today's figures but will change later ones — a submodule pin that drifts silently, and a project reset that adopts leftovers as pristine.
No due date•10/17 issues closedLive credentials reached artifacts on a public repository. This is first because it is the only item with a clock on it that is not ours to control: the exposure continues until the keys are rotated, and nothing else on the roadmap changes that.
No due date•1/1 issues closed