jepa-fx-risk

mathias/jepa-fx-risk

Fork 0

generated from mathias/template-go-web

Files

T

History

mathias c7c182fb4d

CD / Lint / Test / Vet (push) Successful in 5s

Details

fix(context): sync to the estate canonical context-sync.sh, drop Cursor/Aider

Same drift as gitea-mcp: this repo's copy of context-sync.sh was two days
stale against dma-cli's canonical (Mathias's Cursor/Aider decision,
2026-09-01) and CLAUDE.md itself was stale enough to need 97 lines of
correction on regeneration -- this repo's context:sync was apparently never
wired into `task check` at all (no drift assertion existed in its Taskfile),
so nothing had been catching it.

Found while building the estate-wide context-sync drift gate (ADR-0019).

- script synced to canonical hash
- .cursorrules, .aider.conventions.md, .aider.conf.yml deleted
- stale context:sync:cursor Taskfile shortcut removed (would now hit the
  canonical script's own `exit 1` on an unknown adapter name)
- CLAUDE.md, AGENTS.md, .context/system-prompt.txt regenerated against canonical

task check -> exit 0.

2026-09-05 08:01:39 +02:00

autoresearch_start.py

feat(loop): add --run-dir isolation, heartbeat, ntfy-on-crash; scaffold start command

2026-06-27 10:19:17 +02:00

check_gpu.py

chore(phase0): reproducible compute gate — torch cu130 + GPU smoke test

2026-06-23 16:43:01 +02:00

context-sync.sh

fix(context): sync to the estate canonical context-sync.sh, drop Cursor/Aider

2026-09-05 08:01:39 +02:00

fetch_data.py

feat(data): EUR/USD M1 fetch + daily realized-vol prep (toy slice, #2/#11)