kyle 0a3b8d70dc move debug handoff to research ...

maybe not the best place but the closest we have

2026-09-26 01:39:39 -07:00

..

plans

T44: test paths removed by guards; the checklist compares /tmp around a gate run

2026-09-26 01:29:46 -07:00

research

move debug handoff to research

2026-09-26 01:39:39 -07:00

specs

M5b spec: one instance per daemon, by a lock file

2026-09-26 01:34:35 -07:00

decisions.md

decisions: the local-model trailer, and adopting the combined Ornith guidance

2026-09-26 00:39:50 -07:00

dependencies.md

memoryd: the crate, memoryd.toml, and listening on a socket

2026-09-25 16:41:44 -07:00

design.md

Apply P21: identity and experience; memory writes without approval; built-in policy

2026-09-24 13:53:02 -07:00

egress.md

Specify M4a (conversations over Mattermost); propose P15 on secrets

2026-09-23 17:30:31 -07:00

implementer-lessons.md

T44: test paths removed by guards; the checklist compares /tmp around a gate run

2026-09-26 01:29:46 -07:00

implementer-log.md

M5a review: tasks 01 to 22 and the device check; tips I16, I17, T34 to T40

2026-09-25 21:04:23 -07:00

inference-contract.md

Record straylight's new slot layout: four slots over one 262144-token pool

2026-09-22 21:11:23 -07:00

milestones.md

P24: M5b is hardening, M5c Honcho, M5d the scheduler

2026-09-25 23:53:12 -07:00

runbook.md

runbook: the M5a memory entries as the plan gave them

2026-09-25 21:02:32 -07:00