mathias

mathias commented on issue mathias/hyperguild#82

2026-09-23 12:46:12 +00:00

Surface aging brain/raw/ pending notes instead of relying on someone remembering to check

Already shipped 2026-08-18 (commit 3e537bd, "feat(brain): surface an aging review queue in brain_context (#82)") — issue just never got closed alongside it.

Option 2 from this issue's own…

mathias closed issue mathias/hyperguild#82

2026-09-23 12:46:12 +00:00

Surface aging brain/raw/ pending notes instead of relying on someone remembering to check

mathias closed issue mathias/hyperguild#101

2026-09-23 12:44:46 +00:00

cd.yml fails on a duplicate run: "nothing to commit" is treated as an error

mathias commented on issue mathias/hyperguild#101

2026-09-23 12:44:45 +00:00

cd.yml fails on a duplicate run: "nothing to commit" is treated as an error

Duplicate of #118 (same defect, reported independently ~3 weeks apart), which was fixed and closed 2026-09-17 — this issue just never got linked/closed alongside it.

Verified current `.gitea/work…

mathias closed issue mathias/hyperguild#120

2026-09-23 12:44:27 +00:00

Procfile, Taskfile and smoke-routing.sh still invoke cmd/supervisor and cmd/routing, neither of which exists

mathias commented on issue mathias/hyperguild#120

2026-09-23 12:44:25 +00:00

Procfile, Taskfile and smoke-routing.sh still invoke cmd/supervisor and cmd/routing, neither of which exists

Already fixed, just never linked to close this issue. PR #121 (commit 6e449b4, "fix(taskfile): remove build targets for binaries deleted months ago, and make check build") landed all three…

mathias closed issue mathias/hyperguild#134

2026-09-23 12:43:34 +00:00

Live-triage the real brain/raw/ backlog with brain_reject (deferred from #81)

mathias commented on issue mathias/hyperguild#134

2026-09-23 12:43:32 +00:00

Live-triage the real brain/raw/ backlog with brain_reject (deferred from #81)

Executed. brain_reject was still not visible in this Claude Code process's MCP tool list (the harness caches tools/list for the life of the process — a fresh subagent dispatch doesn't get a…

mathias commented on issue mathias/hyperguild#123

2026-09-23 10:28:23 +00:00

Audit remaining write paths that reach disk without claim-envelope validation (brain_edit, ingestion pipeline)

Closed via PR #136 (squash-merged to main).

Summary

brain_edit now validates the claim envelope before writing, same as brain_write/brain_update (#111). Getting there took 7 adversarial…

mathias pushed to main at mathias/hyperguild

2026-09-23 10:28:09 +00:00

e157311cd8 fix(mcp): validate the claim envelope before brain_edit writes too (#136)

mathias closed issue mathias/hyperguild#123

2026-09-23 10:28:08 +00:00

Audit remaining write paths that reach disk without claim-envelope validation (brain_edit, ingestion pipeline)

mathias merged pull request mathias/hyperguild#136

2026-09-23 10:28:08 +00:00

fix(mcp): validate the claim envelope before brain_edit writes too

mathias pushed to fix/123-brain-edit-envelope-check at mathias/hyperguild

2026-09-23 10:25:30 +00:00

442557ca31 docs(mcp): document checkClaimEnvelope's accepted v1 gap

mathias commented on issue mathias/hyperguild#123

2026-09-23 08:55:04 +00:00

Audit remaining write paths that reach disk without claim-envelope validation (brain_edit, ingestion pipeline)

Round 7 (PR #136, commit 42578c2) correctly closes round 6's exact repro — frontmatterScalars now folds an indented continuation line into a top-level key's empty inline value before `ValidateEn…

mathias pushed to fix/123-brain-edit-envelope-check at mathias/hyperguild

2026-09-23 08:51:11 +00:00

42578c21b9 fix(mcp): fold YAML block-scalar continuations before validating the envelope

mathias commented on issue mathias/hyperguild#123

2026-09-23 08:49:12 +00:00

Audit remaining write paths that reach disk without claim-envelope validation (brain_edit, ingestion pipeline)

Round 6 (PR #136, commit 1d67e16) independently re-verified and correctly closes round 5's exact BOM repro.

Verification performed (this pass)

Flip-tested frontmatterBlock's new…

mathias pushed to fix/123-brain-edit-envelope-check at mathias/hyperguild

2026-09-23 08:44:55 +00:00

1d67e16c0d fix(mcp): strip a leading UTF-8 BOM before frontmatter fence detection

mathias commented on issue mathias/hyperguild#123

2026-09-23 08:43:11 +00:00

Audit remaining write paths that reach disk without claim-envelope validation (brain_edit, ingestion pipeline)

Round 5 (PR #136, commit 4f75279) correctly closes round 4's exact CRLF repro — frontmatterBlock now normalizes \r\n→\n once before fence detection, so `TestBrainEdit_RejectsAnInvalidEnvelo…

mathias pushed to fix/123-brain-edit-envelope-check at mathias/hyperguild

2026-09-23 08:40:02 +00:00

4f75279ceb fix(mcp): normalize CRLF before frontmatter fence detection

mathias opened issue mathias/hyperguild#138

2026-09-23 08:39:52 +00:00

Consolidate the three independent frontmatter-fence parsers