• Joined on 2026-03-27
mathias closed issue mathias/hyperguild#22 2026-08-19 18:21:04 +00:00
chore(brain): migrate legacy notes into wing/hall layout
mathias commented on issue mathias/hyperguild#22 2026-08-19 18:21:01 +00:00
chore(brain): migrate legacy notes into wing/hall layout

Verified live against mathias/brain: wiki/concepts/ (20 files) and wiki/entities/ (34 files) are not migration debris — they're active, currently-written wings with their own _index.md,…

mathias commented on issue mathias/hyperguild#36 2026-08-19 18:19:59 +00:00
adr: define stable agent platform interface contract (execute / getEvents / emitEvent / wake / provision)

DECISION NEEDED

Question: Does a stable multi-harness interface contract (execute/getEvents/emitEvent/wake/provision) still make sense now that #75 consolidated to a single harness, or…

mathias commented on issue mathias/hyperguild#29 2026-08-19 18:19:51 +00:00
Track E.3 — skill-from-session generator (BLOCKED on E.1)

No state change — still genuinely blocked on E.1 (the referenced "hyperguild#NN E.1" was never actually filed as far as I can find; worth confirming whether E.1 exists under a different number or…

mathias commented on issue mathias/hyperguild#25 2026-08-19 18:19:48 +00:00
skills(platform): major overhaul — succinct + brain-backed + lazy + reconsider MCP-vs-CLI

Re-scope note per #75 (closed 2026-08-19), not closing this one — most of it still stands:

  • Its blocking gate (#34) is closed as obsolete — the routing-pod pass-rate validation it waited…
mathias closed issue mathias/hyperguild#35 2026-08-19 18:19:36 +00:00
Validate the core hypothesis: MVP pass-rate dataset + fast-model kill-date
mathias commented on issue mathias/hyperguild#35 2026-08-19 18:19:33 +00:00
Validate the core hypothesis: MVP pass-rate dataset + fast-model kill-date

Obsolete per #75 (closed 2026-08-19): same disposition as #34. The core hypothesis this issue asks to validate (a fast local model can substitute for the thinking model on high-pass-rate skills,…

mathias closed issue mathias/hyperguild#34 2026-08-19 18:19:30 +00:00
chore: gate #25 and #29 on routing bootstrap + fast-model validation (30-day hold)
mathias commented on issue mathias/hyperguild#34 2026-08-19 18:19:27 +00:00
chore: gate #25 and #29 on routing bootstrap + fast-model validation (30-day hold)

Obsolete per #75 (closed 2026-08-19): this gate's entire premise — validate the routing pod's fast-model hypothesis via real /pass-rate traffic before starting #25/#29 — no longer applies. #75…

mathias closed issue mathias/hyperguild#32 2026-08-19 18:19:24 +00:00
chore: verify brain injection wired in all four routing skill handlers
mathias commented on issue mathias/hyperguild#32 2026-08-19 18:19:21 +00:00
chore: verify brain injection wired in all four routing skill handlers

Obsolete per #75 (closed 2026-08-19): all four target handlers (internal/skills/{review,debug,retrospective,trainer}/handlers.go) are iceboxed, not part of the live tree — recoverable at tag…

mathias closed issue mathias/hyperguild#31 2026-08-19 18:19:18 +00:00
fix: connect routing-mcp to claude.ai as custom MCP connector (project_create gap)
mathias commented on issue mathias/hyperguild#31 2026-08-19 18:19:14 +00:00
fix: connect routing-mcp to claude.ai as custom MCP connector (project_create gap)

Obsolete per #75 (closed 2026-08-19): cmd/routing — the entire routing-pod binary this issue's project_create connector target lived in — is iceboxed (recoverable at tag `icebox/cmd-routing-2…

mathias closed issue mathias/hyperguild#75 2026-08-19 18:18:19 +00:00
Consolidate to single harness: pin regressions, icebox cmd/routing, confirm hyperguild-alone is sufficient
mathias commented on issue mathias/hyperguild#75 2026-08-19 18:18:16 +00:00
Consolidate to single harness: pin regressions, icebox cmd/routing, confirm hyperguild-alone is sufficient

Verdict: clean-cut

This was actually executed via PR #76 (merged) during a concurrent CAD burst on 2026-08-18, but the required report-back was never posted here — filing it now after…

mathias pushed to main at mathias/cad-atlas 2026-08-19 11:38:55 +00:00
1fdd7d2e3d ci: build via buildctl instead of rootless buildah (infra#241)
mathias pushed to main at mathias/cad-atlas 2026-08-19 11:26:29 +00:00
99fb969d27 fix(test): stop asserting CI job names that the live overlay legitimately replaces
mathias pushed to main at mathias/cad-atlas 2026-08-19 10:00:19 +00:00
294d9e2fc5 feat(atlas): stage 05 reflects the dispatch execution path
mathias closed issue mathias/hyperguild#65 2026-08-19 07:49:42 +00:00
chore: prune merged feature branches (hyperguild + infra)
mathias commented on issue mathias/hyperguild#65 2026-08-19 07:49:36 +00:00
chore: prune merged feature branches (hyperguild + infra)

Same dispatch stall as #54/#4/#39 — falling back to direct execution. Verified all 8 hyperguild branches against pr_list (each has a closed PR into main with a legitimate shipped-feature…