brain-gardener#5: the audit's unsourced check was flagging 59 raw
claudewatcher session dumps as high-severity hallucination risk,
because it can't distinguish a first-party observation (a session
transcript) from an external claim needing citation.
Adds WriteNoteOptions.SourceType, emitted as source_type: <value> in
the wing/hall frontmatter route. claudeSink (the claudewatcher sink)
now always sets source_type: internal on the notes it writes.
Wires the curation primitives into the MCP surface (all three sites:
tools() descriptors, handleCall dispatch, package doc) and registers the
GET /pending + POST /promote REST routes. brain_promote additionally
re-indexes the promoted note into the graph (best-effort), matching the
other write paths. brain_pending is the human-review-queue complement to
the agent-facing brain_write.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Closes the human curation loop: list brain/raw/ notes awaiting review
(oldest-first, with excerpt) and promote one into brain/wiki/<wing>/<hall>/.
PromoteNote rewrites frontmatter (sets wing/hall/promoted_at, preserves
created_at + custom fields via the existing frontmatter editor), rebuilds
the wing _index, and runs auto-tunnel. Atomic from the caller's view:
hall/wing/slug/collision validation happens before any fs change, and the
source is deleted only after the destination write succeeds (write-then-
delete, never move) — a collision or write failure leaves raw/ intact.
Default slug = filename minus the YYYY-MM-DD- prefix.
Also exposes GET /pending + POST /promote for shell scripts (bad
hall/collision/missing-source → 400, not 500).
Scope note: operates on raw/ (the retrospective-skill review queue) per
this issue's spec. The knowledge/ legacy pile is #22's bulk-migration
concern, not this ongoing queue.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A live token-scope probe against mathias/ai-sessions revealed gitea's
contents API uses POST to create and PUT (sha required) to update — the
first impl always PUT'd, so creating a new summary 422'd "[SHA]: Required".
The httptest mock had the same wrong assumption. Pick the method by
whether the file exists (GET sha). Token confirmed contents:write
(push:true) by the probe.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds Client.WriteFile (Gitea contents API) so the Gitea client also
implements capture.SummaryWriter. Upserts: a GET resolves the current
blob sha so an existing file is updated (the richer-fidelity-supersedes
rule for re-captured sessions) rather than 422'd. Owner stays the fixed
const; token only in the Authorization header (no leak — regression
tested). Refactors the HTTP path into a shared request() helper so the
contents flow can branch on 404 without it being an error.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds the `capture` MCP tool: the #55 relay for harnesses that cannot run
the use-case in-process (claude.ai Chat/Cowork/Design, Crush, Pi, LLM
Council). They reach it through the existing /mcp OAuth connector.
- Thin: forwards to the SAME CaptureService as POST /capture; holds no
state and retains nothing beyond the I5 audit record. The containment
properties accepted in infra security-baseline (I2 ledger) hold by
construction.
- Per-principal: ServeHTTP re-derives the caller's principal from the
Bearer header (the chassis middleware gates but discards it) and stashes
it in context; the tool resolves the trust-zone origin from it. A
caller-asserted harness/origin in the body is ignored — origin is
server-derived, so the I1 confidential refusal still fires for us-nexus
callers (claude.ai), and sovereign-allowlisted JWT principals pass.
- Registered only when WithCapture is wired (all three sites: tools(),
handleCall, package doc); main wires REST + MCP from the same service,
resolver, and credentials.
Tests: listed-only-when-wired, forwards-via-static-principal,
confidential-via-us-nexus-refused, confidential-via-sovereign-allowed,
unauthenticated-rejected, caller-cannot-forge-origin.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Lifts the Bearer principal-derivation and the request→CaptureInput decode
out of the REST handler into exported package funcs, so the MCP capture
tool (#55 relay) reuses the exact same auth precedence and wire shape —
one implementation, not two. No behaviour change to POST /capture.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Map capture.ErrAuditUnavailable → HTTP 503 (audit substrate down /
confidential unauditable / floor).
- main: buildAuditSink selects the DegradingSink (loki central + durable
file buffer under brainDir + optional ntfy) when BRAIN_LOKI_URL is set
and starts the reconcile loop; else the plain slog sink. Notifier kept
as a nil interface (not typed-nil) when unconfigured so the sink and
reconcile skip it cleanly.
Env: BRAIN_LOKI_URL, BRAIN_NTFY_URL, BRAIN_NTFY_TOKEN,
BRAIN_AUDIT_RECONCILE_INTERVAL (default 60s). Buffer at
<brain>/.audit-buffer/capture.jsonl.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The classification-aware I5 audit path (§4.4):
- DegradingSink.Reserve: central up → AuditCentral; central down +
confidential → refuse (no buffer); central down + internal/public +
buffer writable → AuditBuffered; central down + buffer unwritable →
floor refuse. Record executes the reserved outcome and, when buffered,
fires an ntfy alert.
- FileBuffer: durable JSONL buffer that survives process restart; Confirm
rewrites the file without a record, so a buffered record is cleared ONLY
after its central write is confirmed.
- LokiCentral: /ready probe + /loki/api/v1/push (full audit entry as the
structured line). NtfyNotifier: degraded-state alerts; token only in the
auth header, never logged (regression-tested).
- Reconcile + StartReconcile: replay buffered records to central on
recovery, confirm-then-clear per record; a failed push keeps the record
buffered (no loss). SlogSink updated to the two-phase shape (always
central, never fails) — the default when no loki endpoint is set.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Splits the audit port into Reserve (before any write) + Record (after),
so "confidential + sink-down → refuse before any write" is literally true
even though the audit record — which lists what landed — can only be
written afterwards.
- AuditSink.Reserve(ctx, level) → AuditOutcome | error. The error path
refuses the capture before writing: confidential + central sink down,
or the all-tiers floor (nothing can record).
- AuditSink.Record(ctx, entry, outcome) persists per the reserved outcome.
- Service: I5 gate runs after the I1 gate and after the dry-run
short-circuit (dry-run never probes the sink). AuditBuffered surfaces on
the receipt. New ErrAuditUnavailable sentinel (→ HTTP 503).
The tier→behaviour decision lives in the sink impl (#54's DegradingSink),
not the service — the service just honours Reserve's verdict.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The HTTP door for the capture capability. Thin: authenticate → derive
trust-zone origin → decode → capture.Service → map receipt to status.
- Auth mirrors the chassis Bearer precedence (static token wins, then Dex
JWT) but returns the resolved principal + auth path, which the chassis
middleware hides — capture needs the principal to derive the origin.
Depends on a small Validator interface (the chassis *JWTValidator
satisfies it) so the JWT/origin path is testable without a live JWKS.
- OriginResolver maps principal → trust zone: static-token caller and
allowlisted JWT subjects → sovereign; every other principal → us-nexus
(fail safe, so the I1 gate refuses confidential by default). Principal
and origin are server-set on the input, overwriting any body the caller
sent.
- HTTP status: 200 all-ok / dry-run, 207 partial, 502 all-failed, 403 on
the I1 refusal, 400 on fail-closed validation.
- Wired in main behind the same static+JWT credentials as /mcp, reusing
the MCP server's graph-wired brain store (one implementation) and the
classification tags (#50). Mounts only when a Gitea tracker is
configured. Sovereign JWT principals via BRAIN_CAPTURE_SOVEREIGN_PRINCIPALS.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Emits the request-level audit record to structured logs (scraped by the
existing alloy/loki substrate) and surfaces security events at warn
level. Placeholder behind the AuditSink interface — the classification-
aware loki+buffer+reconcile sink (confidential fails closed, internal
degrades) lands in #54 and replaces this without touching callers.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds the trust-zone Origin to CaptureContext and the I1 gate to the
use-case: a confidential effective classification through a us-nexus
origin is refused before ANY write (ErrSovereigntyRefused), and the
refusal is itself audited. A caller-asserted harness label that names a
different zone than the server-derived origin is logged as a security
event — context.Harness is descriptive-only, never a gate input.
The gate triggers only on an explicit ZoneUSNexus, so the unset default
(ZoneUnknown) can never make it fire on caller-controllable input; the
REST adapter always sets a concrete zone.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements the IssueTracker port as a real Gitea REST client (#49c) — the
new outbound dependency the brain server gains for capture.
- CreateIssue / CommentIssue / CloseIssue(+optional closing comment) over
the Gitea API. Owner is the const "mathias", never caller-supplied, so
a caller cannot redirect a write to another owner's repo.
- Token read once at construction (BRAIN_GITEA_TOKEN), held in the struct,
travels only in the Authorization header — never logged or in argv.
Error messages carry status + truncated body, never the token
(regression-tested). gitea.New returns nil when URL or token is unset,
so missing config = tracker disabled via one nil check.
- Injected into the MCP server behind the capture.IssueTracker interface
via WithIssueTracker (constructor injection, swappable/testable); main
wires it from BRAIN_GITEA_URL (default https://git.d-ma.be) +
BRAIN_GITEA_TOKEN. Consumed by the capture use-case in #53.
Tests use httptest transports: create (owner+auth header asserted),
comment, close with/without comment, error path that proves the token
never leaks into an error string.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
#52's IssueTracker spec is CloseIssue(repo, number, comment). Refine the
#51 port signature to match and have the service pass the ticket body as
the closing comment (empty ⇒ close only). Keeps the close-with-comment
flow first-class rather than forcing two separate ticket items.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Extracts the #45 write/update/get logic + the wiki upkeep that must
accompany a write (wing _index rebuild, cross-wing auto-tunnel, graph
re-index) into a single concrete brainstore.Store implementing
capture.BrainStore. The MCP brain_write/brain_update/brain_get handlers
are re-pointed at it, so there is one implementation, not two — the DRY
payoff #51 is named for. capture and MCP now share the exact same brain
write path and read-after-write contract.
The Server gains a *brainstore.Store, constructed in NewServer and given
the graph store in WithGraph. Embedding refresh stays out-of-band
(mtime-driven vectorstore.Sync), unchanged. Existing MCP brain_update/
brain_get/brain_write tests pass unmodified — behaviour and the
{id, path, content_hash} response contract are preserved.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Clean-Architecture core of the capture capability (#49b). Pure
orchestration over ports — no HTTP, no live Gitea, no audit I/O — fully
unit-tested against fakes before any adapter exists.
- Ports: BrainStore (#45 write/update/get), IssueTracker, SummaryWriter,
ClassificationPolicy (satisfied by #50's classification.Config),
AuditSink. Entities: Insight, Ticket, Summary, CaptureContext,
CaptureInput, CaptureReceipt.
- CaptureService.Capture: validate-before-write (fail-closed), resolve
effective classification (stricter of declared vs target-derived;
under-declaration logged as a security event), orchestrate insights
(write/supersede) → tickets → summary best-effort, emit a request-level
audit record of exactly what landed, return a partial-aware receipt.
- dry_run short-circuits after validation, writes nothing (not even audit).
Out of scope here, layered on later: the I1 origin sovereignty gate (#53,
needs the server-derived principal) and the classification-aware audit
degradation/refusal (#54). "Effective" is folded into the service as
classification.Stricter rather than a port method — the stricter-wins
rule is use-case policy.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements the I1-prerequisite from #49/#50: the classification taxonomy
and the per-wing / per-repo tagging the capture server reads to derive a
target's sensitivity.
- Levels public < internal < confidential, ordered so "stricter wins"
(spec §4.1 model C) is a plain max via Stricter().
- Tags read from an optional classification.yaml at the brain root
(wings:/repos: maps). Absent file → defaults-only, not an error.
- Defaulting: client-* → confidential; hyperguild/homelab → internal;
everything else → confidential. Fail-safe-to-strictest is the
load-bearing property: a missing tag never silently downgrades.
- Config.Derive(Target) is the function the use-case calls; Wing/Repo
are the per-kind helpers. ParseLevel rejects unknown tokens; a bad
level in the config file is a hard load error.
Central classification.yaml (not _index.md frontmatter, not gitea repo
topics): classifying a repo needs no live Gitea client, so #50 has no
dependency on the tracker work (#52); it's auditable in one place; and
it avoids BuildWingIndex clobbering a wing's regenerated _index.md.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Wires the #45 verbs into the MCP surface (all three sites: tools()
descriptors, handleCall dispatch, package doc comment).
- brain_update: supersede-by-slug or full path; rebuilds wing _index and
re-tunnels cross-wing matches against the new body (idempotent,
best-effort), re-indexes the graph, returns {id, path, content_hash,
superseded}.
- brain_get: fetch by id or path (both are the brain-relative handle);
returns {id, path, content_hash, frontmatter, body}.
- brain_write: return contract extended from {path} to {id, path,
content_hash} — path kept for backward compat — so the create path
also yields a stable handle.
id == relPath; content_hash == sha256 of the file bytes. Tests cover the
supersede happy path, missing-target error + no-create, get by id/path,
write handle, and an end-to-end re-embed test that drives the real
vectorstore.Sync re-index after an update.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements the api-layer half of #45. UpdateNote supersedes a note in
place (whole-note body replace, frontmatter re-stamp: updated_at,
supersedes=prior content hash, supersede_reason), preserving created_at,
wing, hall, and any custom fields. Never creates — a missing target is
an error so callers fall back to brain_write. ReadNote is the read-after-
write primitive (frontmatter + body + content_hash). ContentHash is the
sha256 handle that round-trips write/update → get.
Frontmatter is edited via a line-preserving ordered editor rather than a
yaml.v3 round-trip, which would reorder keys and strip comments — the
brain writes flat key:value frontmatter by hand.
Embeddings are not refreshed here: the rewritten file's mtime advances,
which the out-of-band vectorstore.Sync ticker uses to re-embed it — the
same mechanism brain_write relies on.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A doc with no headings and no blank-line paragraphs (JSON-lines, e.g.
wiki/telos/decisions/human-intent-column.md) survived both chunk passes whole
and was sent to nomic-embed over its context window → 'input length exceeds
the context length' (400, the steady embed errors=1). Add a final hard-split
pass (line then UTF-8 rune boundaries) so no chunk exceeds maxBytes. TDD.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Qwen3-Reranker is a web-search cross-encoder. Against conversational /
personal-intent queries (e.g. 'what am I optimizing toward?') it scores every
candidate as 'no', so brain_answer collapsed to 'No relevant content found'
even though BM25 had retrieved on-topic notes (incl. the telos wing). Treat
the reranker as a filter: when it keeps zero results, fall back to the
BM25/vector ordering (capped to the no-reranker depth of 10). Refs brain#11.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A ~/.zshrc read surfaced OP_SERVICE_ACCOUNT_TOKEN into a transcript. The
_TOKEN= form was already caught by homelab-env-token, but a bare ops_<b64>
value was not. Add an op-service-account rule (ordered early). Tests cover
both env-assigned and bare forms.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A shell mangle that glued a key to a preceding word ('yes'+'sk-...') had
no word boundary, so the leading \b in the openai-sk rule failed to match
and a LiteLLM master key leaked past the scrubber into ingest (2026-06-11).
- openai-sk: drop leading \b, match sk- shape anywhere ({32,} floor keeps
short task-/disk- words clean).
- add jwt rule for bare header.payload.sig tokens (no Bearer prefix).
- regression tests: the exact yessk- evasion, standalone sk-, bare JWT,
plus clean-content guards for task-/disk-.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pre-rollout guard. Source code stays clean — client identities come
from CLAUDE_INGEST_CLIENT_BLOCK env (sourced from a SOPS-encrypted k8s
secret in infra repo). Env value is a regex alternation; main wraps
it with `(?i)\b(...)\b` so word-boundary matching avoids false hits
inside longer identifiers (e.g. "Sebastian" doesn't trigger on "SEB").
DefaultRules (credential shapes) still take precedence so any leak
that's BOTH a client mention AND a credential shape logs as the
credential — strictly more dangerous, points triage at the right
thing. Tests cover precedence + case variations + word-boundary
respect + invalid-pattern rejection.
Refs: infra#73 Track E.1 pre-rollout grill (option B).
Bump-Type: minor
New package internal/claudewatcher. The volume gate (24 turns/week of
agentsquad logs vs 500/week gate) exposed that the real signal lives
in daily Claude Code usage at ~/.claude/projects/*/<uuid>.jsonl, not
in agentsquad output. This package captures that signal. See infra#73
Track E + hyperguild#27 for the full reframe.
Components:
- parser: tolerant JSONL parser over the observed Claude Code session
schema (user / assistant / attachment / system + bookkeeping types).
Skip-flag fast-paths queue-operation, last-prompt, permission-mode,
ai-title, bridge-session, file-history-snapshot.
- scrubber: 11-rule fail-closed regex set for credential shapes
(bearer, postgres URIs, PEM, ssh-key, ghp_/sk-/sk-ant-/AKIA, homelab
env tokens, SOPS markers). Drop turn + log on match.
- cursor: postgres-backed claude_session_cursors table, keyed by
(host, file_path) with byte_offset. Resumable across pod restarts.
- watcher: poll loop. Walks SessionsDir, processes each .jsonl from
its cursor offset, runs scrubber, emits a Batch per file to a
Sink interface, advances cursor on successful Ingest.
No classifier integration in this commit — every kept turn is emitted
in a per-session batch. The cmd/server wiring (next commit) routes
batches to brain/wiki/claude-sessions/facts/. Classifier-driven hall
routing (decisions / failures / hypotheses) is a follow-up.
19 unit tests across parser + scrubber + watcher. task check green.
Refs: infra#73, hyperguild#27
Returns top-N relevant brain entries for a project context. Combines
BM25 hits on project name with 2-hop graph expansion via Track A's
graphstore (when BRAIN_GRAPH_ENABLED). Closes hyperguild#28.
Notes on implementation choices that deviate slightly from the spec:
- Excerpt length: 200 chars per spec (vs the 300 used by search.Result).
truncateExcerpt clamps the already-stripped BM25 excerpt; graph-only
neighbours load their excerpt from disk via a private readExcerpt
helper (search.hydrate is unexported).
- Graph scoring: 0.6 / max(1, distance) per neighbour, so distance-1
contributes 0.6 and distance-2 contributes 0.3. BM25 hits decay
linearly from 3.0 (rank-0) to 1.0 (rank-2), giving BM25 hits a
natural ceiling above pure-graph hits while still letting a doc
surfaced via both edge types outrank a BM25-only one.
- Test placement: package mcp (internal) rather than mcp_test, because
graphReader is unexported and WithGraph only accepts *PGStore; an
internal test can install a dual-interface fake directly on s.graph
without spinning up postgres.
Bump-Type: minor
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Initial M4 mapping put wiki/entities/* in tier=note. Post-M4 eval
regressed qwen35-9b-fast from rank 2 → off top-5: knowledge entries
that cite the entity in passing now outscore the entity page itself
(1.5× weight vs 1.0×).
Entity anchor pages are durable facts about concrete things — they
map cleanly to the knowledge/facts/ slot in the post-M3 layout
target. Promote them now so the path inference matches.
Eval re-run after deploy is in infra#72.
The eval set under brain/eval/qa-2026-05.md showed BM25 top-1 at 20%
with 5 of the missing slugs being short focused knowledge entries
that lost to long aggregate docs on raw term-frequency. Tier weighting
addresses that without touching the BM25 algorithm itself.
How
- Result struct gains a Tier field, populated during the file walk
via extractTier (frontmatter wins, path prefix as fallback —
mirrors the graph.inferTierFromPath logic so the two callers stay
in lockstep).
- After the existing sort (and optional hybridMerge), do a final
stable re-sort by float64(Score) * tierWeight(Tier). Knowledge
×1.5, note ×1.0, inbox ×0.3, unknown ×1.0.
- hydrate() (vector-only hits) also fills Tier so re-ranking covers
the hybrid path.
Test covers the load-bearing case: a long note-tier doc with raw=10
loses to a short knowledge-tier doc with raw=8 after weighting
(8×1.5=12 vs 10×1.0=10).
Measurement gate is in infra#72: re-run brain/eval/score.py against
the live brain after this image lands; close the issue when top-1
hit rate lifts by ≥10 absolute points.
extract.go now reads `tier:` and `topic:` from YAML frontmatter, with
a path-based fallback when frontmatter is absent (the pre-M3 state on
every existing entry):
knowledge/* → tier=knowledge
notes/* → tier=note
wiki/** → tier=note (sources + concepts + entities are I-level)
inbox/**, raw/**, sessions/**, clips/** → tier=inbox
Frontmatter wins when present — covers the M3-migrated case where an
entry's path may not match the tier the author chose for it.
UpsertEntity persists both columns. M1's schema already has them.
Backfill on next pod start populates tier for the whole corpus
without any file moves; M3 will follow up with the actual layout
migration and explicit frontmatter writes.
Schema-only change. DDL adds tier + topic on fresh tables and uses
ADD COLUMN IF NOT EXISTS on existing tables (idempotent across pod
restarts). New conditional indexes match the wing/hall pattern.
No behavior change in this commit — UpsertEntity still writes only
the original columns; tier + topic stay '' on every row. M2 plumbs
the parser through. The empty default means existing queries are
untouched until the rest of the chain lands.
Part of infra#72 — brain DIKW tier redesign.
classifyByPath had a hole: paths like wiki/index.md or wiki/<slug>.md
(direct children of wiki/, no subdirectory) hit the default branch and
wrote Wing=parts[1] — which IS the filename, not a wing. Symptom in
brain_entities: rows like (slug=index, wing=index.md) and
(slug=autobe-..., wing=autobe-evaluation-pattern-....md).
Fix: when len(parts) < 3 (no subdirectory at all), fall through to
Type=knowledge and let frontmatter set wing/hall if present.
Add brain/eval/ artifacts at the same time:
- qa-2026-05.md — 20 hand-authored Q→expected-slug pairs covering the
homelab knowledge corpus across mcp, dex, gitops, postgres, go,
models, methodology
- score.py — calls brain_query for each pair, scores top-1 + top-3,
emits per-question detail. BRAIN_MCP_TOKEN via env.
Pre-fix baseline against the live brain: top-1 = 20% (4/20),
top-3 = 65% (13/20). Six hard misses where the expected slug doesn't
even land in the top-5.
Used to gate the phase 2 DIKW redesign (infra#62 follow-up): if
phase 1 fixes (this parser fix + 20 backlink authoring on top
orphans) lift top-1 by <10 absolute points, structure is the
bottleneck and the tier redesign is justified.
Commit 4 of Track A — the no-shelfware close-out the grill demanded.
brain_answer now folds the 1-hop outgoing neighbourhood of its top
BM25/rerank hit into the LLM's context as a <related> block when
BRAIN_GRAPH_ENABLED is on. With the flag off the prompt is byte-for-
byte identical to the pre-Track-A behaviour, so existing tests still
pass without modification.
The hop list contains slug, edge_type, doc_path — no extra retrieval
pass, no second LLM call, no file reads. The model can ignore the
block when irrelevant; when it adds signal we get GraphRAG for free.
Refs: docs/superpowers/specs/2026-05-homelab-training-graph-next-step.md
in infra repo + grill addendum item "Track A: GraphRAG wiring into
brain_answer is mandatory in same commit chain (no shelfware risk)".
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Commit 3 of Track A. The MCP server now publishes a new tool that
opens the brain knowledge graph (entities + wikilink edges) for
external consumers (claude.ai connectors, gitea-mcp, agentsquad).
- tools_graph.go: brain_graph handler dispatches by op:
neighbors — 1-hop outgoing from slug, optional edge_type filter
subgraph — every reachable slug within depth hops (≤6)
path — shortest directed path src→dst within depth (≤8)
Returns slug + entity metadata + edge_type + hop distance.
- server.go: handleCall routes "brain_graph" to brainGraph.
- handlers.go: tool descriptor with the op enum + per-op required
fields documented in the description.
- server_test.go: TestServerToolsList expects brain_graph in the
listing.
The tool returns an error when BRAIN_GRAPH_ENABLED is unset — same
shape as brain_answer when the answer LLM is unconfigured.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Commit 2 of Track A. Service stays a no-op until BRAIN_GRAPH_ENABLED=
true; flipping it on creates the schema (idempotent), starts indexing
every successful write, and optionally backfills the existing brain
dir.
- internal/graphsync: best-effort wrapper around graph.Extract +
graphstore. IndexDoc reads docPath under brainDir, parses, upserts
entity + replaces edges. BackfillFromBrainDir walks wiki/ +
knowledge/. Both are no-ops on nil store so callers wire
unconditionally.
- mcp.Server gains WithGraph builder + graphsync.Store field.
brain_write, brain_ingest, brain_ingest_raw, brain_tunnel call
indexInGraph after success — failures slog.Warn but never
propagate (graph is augmentation, not correctness).
- cmd/server gates the wiring on BRAIN_GRAPH_ENABLED=true (default
off so first rollout doesn't surprise). BRAIN_GRAPH_BACKFILL=true
triggers a one-shot walk of the brain dir on boot.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Foundation for Track A (GraphRAG on top of existing wiki). Two new
packages, both unwired — service behaviour unchanged until commit 2
hooks the pipeline.
- internal/graph: pure parser. Extract() walks markdown + frontmatter
and emits one Entity + N wikilink Edges per doc. Dedupes per (dst,
line), ignores self-references, classifies hall/concept/entity/
source/knowledge from path layout.
- internal/graphstore: pgx-backed PGStore mirroring vectorstore's
shape. Idempotent Init() creates brain_entities + brain_edges with
indexes on src_slug, dst_slug, src_doc, wing, type. Operations:
UpsertEntity, ReplaceEdgesForDoc (tx), DeleteByDoc, Neighbors,
Subgraph (recursive CTE, depth ≤6), Path (shortest path, depth ≤8).
Schema lives on the shared postgres18 instance alongside the
brain_embeddings table — no new datastore. See
docs/superpowers/specs/2026-05-homelab-training-graph-next-step.md
in infra repo + infra#62.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Second port of the MCP chassis (gitea-mcp was first, commit 658f4ba).
Closes the chassis-adoption loop on the two highest-LOC consumers.
Changes:
- Drop ingestion/internal/auth/ entirely (jwt.go + jwt_test.go +
protected_resource.go + protected_resource_test.go) — chassis provides
JWTValidator + ProtectedResourceHandler with identical semantics.
- Drop ingestion/internal/mcp/auth.go (BearerAuth function, ~65 LOC)
and the integration test auth_test.go (~200 LOC) — chassis
BearerMiddleware replaces it. Static-Bearer-or-Dex-JWT precedence and
RFC 9728 resource_metadata challenge behavior preserved 1:1.
- cmd/server/main.go: import chassis as `chassisauth`, rewire the three
call sites. Use realm="brain" in the BearerMiddleware call so a 401
challenge identifies the resource as the brain MCP.
OAuth client_credentials handler (ingestion/internal/oauth) stays —
chassis v0.1.0 covers only the JWT path; OAuth flow is a candidate for
chassis v0.2.0 once a second MCP needs it (rule of three).
Net delta: -~330 LOC of duplicated auth code; +1 import; +1 GOPRIVATE
env requirement on dev machines (documented in the spike handoff
2026-05-22-mcp-chassis-spike.md).
task check green (lint + test + vet + govulncheck).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Closes infra#50.
Adds an internal/metrics package with a hand-rolled Prometheus
exposition layer (stdlib + sync/atomic only — no new dep) and wraps the
HTTP mux with a timing middleware. Every request emits one observation
on the `brain_query_duration_seconds` histogram labeled by
`path` (request Pattern, low cardinality) and `status` (2xx/3xx/4xx/5xx).
Dependency choice: hand-rolled rather than github.com/prometheus/client_golang
because the surface needed is small (one histogram + bucket constants)
and the repo CLAUDE.md keeps deps stdlib + jwx + testify only. ~150 LOC
of code + tests is cheaper than the chart of transitive prometheus deps.
Endpoints:
- GET /metrics — OpenMetrics text exposition, no auth (cluster-internal)
Wire format pinned by tests in internal/metrics/metrics_test.go. The
ServiceMonitor that drives the kube-prometheus-stack scrape lives in
infra/k3s/apps/supervisor/ (separate commit on mathias/infra).
After this image deploys, the canary alert from
docs/superpowers/specs/2026-05-homelab-architecture-review.md becomes
wireable:
histogram_quantile(0.95,
sum(rate(brain_query_duration_seconds_bucket[5m])) by (le))
> 1.5 * histogram_quantile(0.95,
sum(rate(brain_query_duration_seconds_bucket[5m] offset 7d)) by (le))
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Removes the TODO in Sync that left files static after their first embed.
Edits to brain/wiki/ and brain/knowledge/ now surface in subsequent
syncs without manual /backfill-embeddings calls.
Approach
- Store interface: KnownPaths → KnownPathsWithTime returning path →
updated_at. Callers compare against file mtime to detect edits.
- PGStore: SELECT path, updated_at FROM brain_embeddings.
- Sync groups known chunks by parent path and tracks the EARLIEST
updated_at per parent. A file is stale when its mtime is after that
oldest chunk's timestamp — any chunk older than the file means at
least one chunk hasn't been refreshed since the last edit.
- Stale-path rewrite: delete every old chunk for the parent (handles
"file shrunk → fewer chunks → orphan rows at higher #NNNN" cleanly),
then re-chunk + re-embed + re-upsert.
Tests
- New: TestSync_ReembedsFileWhenMtimeNewer — file mtime forced into the
future vs store updated_at; Sync deletes old chunk + upserts fresh one.
- New: TestSync_SkipsFileWhenMtimeOlder — file mtime backdated; Sync is
a no-op (no upserts, no deletes).
- Updated: stubStore.known is now map[string]time.Time. A zero value
resolves to a far-future sentinel so existing "skip if already known"
tests keep passing without per-test setup.
- pg_test renamed KnownPaths integration → KnownPathsWithTime; asserts
updated_at is non-zero and within 5s of insert wall-clock.
Backward compat
- brain_embeddings rows pre-dating this change carry valid updated_at
values (column was always populated via `DEFAULT now()` + ON CONFLICT
`updated_at = now()`). No migration needed. Live pod will start
re-embedding any file whose source has been edited since its chunks
were originally written.
Closes gitea/mathias/hyperguild#23.
Long markdown files (>~8KB) silently failed to embed because nomic-embed-text
on iguana has a 2048-token context. embed sync logged errors=1 every cycle
with no useful body until #37 added per-item logging — three files exceed
the ceiling: finbert source (8 KB), koala-machine-state (7.1 KB),
litellm-absorption (8.8 KB). Curated knowledge entries should never be
vector-blind.
Approach: chunk-before-embed, no schema change.
vectorstore/chunk.go (new)
- ChunkMarkdown splits at H1/H2 boundaries; sections over maxBytes are
further split at paragraph boundaries, packing greedily under budget.
- NumberChunks assigns "<parent>#NNNN" storage paths (1-based, zero-padded
to 4 digits — handles files with up to ~10k sections in stable sort order).
- ParentPath strips the chunk suffix for retrieval-side dedup.
vectorstore/sync.go
- After ChunkMarkdown produces N pieces, each is embedded + upserted as a
separate brain_embeddings row at "<parent>#NNNN". maxChunkBytes = 4000
(≈1000 nomic tokens, well under the 2048 ceiling with headroom for
unicode/code blocks).
- "Already embedded?" check now reduces known paths to parent set via
ParentPath, so the first chunk hit short-circuits the file.
- Delete walk also reduces via ParentPath; when a parent file disappears,
every chunk row (and any pre-existing bare-path row, for backward
compatibility with rows written before this change) gets dropped.
search/search.go
- hybridMerge collapses chunk-path vector hits to parent via ParentPath
before scope check, RRF accumulation, and hydration. A file with three
chunk hits returns one result row, not three.
Backward compatibility: pre-existing bare-path rows in brain_embeddings
keep working — ParentPath returns them unchanged, knownParents handles
them as if they were "wiki/foo.md#NNNN" hits, sync skips re-embed, and
search dedup is a no-op for them. No migration required to ship.
Tests:
- chunk_test.go covers short / heading split / oversized section /
content preservation / chunk numbering / parent-path stripping.
- sync_test.go adds long-file chunking, single-chunk-row short file,
skip-if-any-chunk-known, delete-all-chunks-of-disappeared-file.
Existing tests updated for #NNNN paths.
- search_test.go adds chunk-paths-dedupe-to-parent.
Closes gitea/mathias/infra#38.
The embed sync goroutine only walked brain/wiki/. brain/knowledge/ (112
curated entries, per CLAUDE.md the most-important brain content) had zero
coverage in brain_embeddings — vector retrieval was blind to it. Hybrid
BM25 + pgvector retrieval would never surface a curated knowledge entry
via the vector arm.
Extract the per-root walk into a loop over a small subdir list and add
"knowledge" alongside "wiki". scanDirs is package-level so it stays a
single source of truth for what gets embedded.
Also log each failing item's path + error string from StartSync.
Previously only the aggregate count was logged, so a persistent
`errors=1` per cycle was opaque. With per-item warnings, the actual
ollama "input length exceeds the context length" surface immediately.
Refs gitea/mathias/infra#37 (this commit covers the knowledge/ scan
bug; the long-file chunking bug is a separate change.)
Wires nomic-embed-text (iguana ollama) + pgvector on the shared
postgres18 into brain_query / brain_answer via Reciprocal Rank Fusion.
Pure BM25 stays the default; setting BRAIN_PG_DSN and BRAIN_EMBED_URL
together opts in. Setting one without the other is misconfiguration →
exit 1.
New packages:
- internal/embed
Client.Embed(ctx, text) → []float32 via POST {URL}/api/embed.
Defaults to nomic-embed-text:latest (768 dim). nil-on-empty-URL so
callers gate on a single nil check.
- internal/vectorstore
PGStore wraps a pgxpool against postgres18. Init creates
brain_embeddings(path PK, vector(768), updated_at) + HNSW cosine
index idempotently. Upsert / Delete / Search / KnownPaths.
Sync(brainDir, store, embedder) diffs brain/wiki/ against the store
and upserts new files / deletes removed ones; StartSync runs it on
a ticker (default 300s). Integration tests gated by BRAIN_PG_TEST_DSN.
- scripts/brain-embeddings-init.sql
One-time DBA setup: brain DB, brain_app role, vector extension,
GRANTs. Idempotent.
Search layer:
- search.QueryOptions gains Vector + Embedder fields.
- QueryContext is the cancellable variant; Query stays for callers.
- When both are set, BM25 (top-N) and pgvector (top-4N) candidates
merge via Reciprocal Rank Fusion (k=60, Cormack et al. 2009 — no
tuning knob, robust to scale differences between rankers).
- Vector-only hits are hydrated from disk so callers see uniform
Result records (path, title, excerpt, wing, hall, score).
- Wing/hall filters still apply to vector candidates via path-prefix.
- On embedder/vector errors the search falls back to BM25 — embedding
outage degrades quality but doesn't take the brain offline.
MCP wiring:
- mcp.Server.WithHybridRetrieval(v, e) opt-in setter, same shape as
WithReranker.
- brainQuery and brainAnswer pass the wired vector/embedder through
to search.QueryContext.
REST:
- POST /backfill-embeddings drives Sync synchronously. Returns
{added, deleted, errors[]}. 503 when feature is unconfigured.
cmd/server/main.go:
- BRAIN_PG_DSN + BRAIN_EMBED_URL together enable hybrid; one alone
→ exit 1.
- vectorAdapter bridges *PGStore (returns []Hit) to
search.VectorSearcher (which takes []VectorHit) without either
package importing the other.
- BRAIN_EMBED_SYNC_INTERVAL (default 300s) controls the background
Sync ticker.
Backend pivot from Qdrant to pgvector recorded in DECISIONS.md
2026-05-18 (supersedes 2026-04-08): postgres18 already runs in
databases/ ns, Qdrant was never deployed, one engine beats two.
Dependency: github.com/jackc/pgx/v5 — modern, native pgvector via
parametric vector literals.
Tests:
- embed.Client: empty-URL nil, request shape, dimension, upstream
error propagation, empty-text rejection.
- vectorstore.PGStore: dimension validation (unit); upsert/search/
KnownPaths (integration, BRAIN_PG_TEST_DSN-gated).
- vectorstore.Sync: adds new files, skips known, deletes
disappeared, skips _index.md, no-op when nil, collects embedder
errors.
- search.Query: hybrid promotes vector-only hits via RRF; falls
back to BM25 on embedder error.
Closes hyperguild#8.
Adds an opt-in cross-encoder rerank step between BM25 retrieval and LLM
synthesis. With BRAIN_RERANKER_URL set, brain_answer retrieves BM25
top-20, scores each excerpt against the query via Qwen3-Reranker on
Ollama, drops the "no" answers, and forwards up to 5 surviving sources
to the LLM. Unset, behaviour is unchanged (BM25 top-10 → LLM).
The reranker is a *filter*, not a re-ranker: Qwen3-Reranker emits a
binary yes/no token under its native chat template, and ties within the
"yes" set are broken by BM25 rank — what got retrieved first stays
ahead.
New package ingestion/internal/reranker:
- Client with URL, Model, HTTP fields.
- New(url, model) returns nil on empty url so callers can treat
"feature disabled" as a single nil check.
- Score(ctx, query, docs) issues one /api/generate call per doc using
the Qwen3-Reranker yes/no chat template (verbatim, because the model
was trained on this exact wording). Parses the first non-think token.
Wiring:
- mcp.Server gains a WithReranker fluent setter to keep NewServer
signature stable.
- brain_answer's BM25 limit jumps to 20 only when a reranker is wired,
to give the filter something to do.
- cmd/server/main.go reads BRAIN_RERANKER_URL (+ optional
BRAIN_RERANKER_MODEL, default dengcao/Qwen3-Reranker-0.6B:F16).
Tests cover: nil-on-empty-url, ordered yes/no scoring, request shape
(model, prompt contents, yes/no template), ambiguous response → 0,
empty doc slice, upstream-error propagation, plus an end-to-end
brain_answer integration that proves only the relevant note reaches the
LLM when noise.md is rejected.
Closes hyperguild#7.
Adds a minimal RFC 8414 + RFC 6749 client_credentials flow so claude.ai's
custom-MCP integration (no static-Bearer field in the UI) can exchange a
client_id + client_secret pair for the existing BRAIN_MCP_TOKEN and use
it as a Bearer on /mcp. No JWTs, no refresh, no expiry — the rest of
the auth middleware is unchanged.
New package ingestion/internal/oauth:
- MetadataHandler(issuer): serves /.well-known/oauth-authorization-server
with grant_types=[client_credentials] and both
token_endpoint_auth_methods (post + basic).
- TokenHandler(cfg): serves /oauth/token. Validates client_id and
client_secret via constant-time compare; returns BRAIN_MCP_TOKEN as
access_token. RFC 6749 §5.2 error JSON on bad grant / bad creds.
Wiring in cmd/server/main.go: opt-in by setting both OAUTH_CLIENT_ID and
OAUTH_CLIENT_SECRET. Setting only one is misconfiguration → exit 1.
Mounts both endpoints with no auth; MCP_RESOURCE_URL supplies the
issuer.
Also pivots issue #8's vector backend from Qdrant to pgvector (see
DECISIONS.md 2026-05-18) — Qdrant was never deployed and postgres18 with
pgvector already runs as the project default; supersedes 2026-04-08 for
this use case.
Tests cover post-auth, basic-auth, wrong secret, bad grant, GET
rejection, malformed Basic header, and Basic without colon.
Closes hyperguild#5.
Adds the `brain_tunnel` MCP tool and auto-tunnel behaviour for
`brain_write`, so concepts that appear in multiple wings become
navigable from any of them.
New surface in package brain:
- WriteTunnel(brainDir, src, tgt) — appends a `## See also` bidirectional
wikilink between two notes in different wings. Idempotent (link not
duplicated on re-call) and reuses an existing See also section.
- DetectTunnels(brainDir, content) — walks brain/wiki/, returns
TunnelCandidates for notes whose title appears in content. Tags
whole-word case-insensitive hits as Exact=true and substring-only hits
as Exact=false.
- AutoTunnel(brainDir, src, content) — wraps DetectTunnels: writes
cross-wing exact matches, stages fuzzy matches into
brain/raw/tunnel-candidates-<YYYY-MM-DD>.md for human review.
MCP wiring:
- `brain_tunnel` tool: explicit manual link (source, target).
- `brain_write` with wing+hall now triggers AutoTunnel on the new
content. Failures are logged and never abort the primary write.
readTitleAndCreated also humanises the slug fallback (hyphens → spaces)
so titleless notes participate in content matching.
Closes hyperguild#16.
Tests: idempotency, same-wing rejection, missing-note rejection,
See-also reuse, exact/fuzzy detection, slug fallback, MCP tool happy
path, auto-tunnel hook (cross-wing exact → linked; same-wing → skipped;
fuzzy → candidates file).
Reorders BearerAuth so a valid BRAIN_MCP_TOKEN match wins instantly and
never emits WWW-Authenticate. Adds RFC 9728 resource_metadata challenge
header on 401 (only when MCP_RESOURCE_URL is configured) so claude.ai's
OAuth-discovery path still works.
Why: claude CLI on koala/flamingo with `.mcp.json` `Authorization: Bearer
$BRAIN_MCP_TOKEN` was being kicked into RFC 7591 dynamic client
registration against Dex (static-only) and dying. Cause was the auth
middleware running JWT validation first and emitting an OAuth challenge
on the fall-through 401 even when the caller had a valid static token.
Inverting the precedence and gating the challenge on resourceMetadataURL
keeps the LAN/Tailscale CLI path silent and only invites OAuth discovery
on actually-unauthenticated requests.
Regression guards in the test file:
- valid static Bearer 200 has no WWW-Authenticate
- 401 with resourceMetadataURL set carries the challenge
- 401 with empty resourceMetadataURL emits no challenge
Closes hyperguild#9 in code. Live verification (claude CLI on koala
listing brain tools) blocked on ingestion image rebuild + redeploy.
Adds a two-dimensional address (wing, hall) to brain notes. A wing is a
topic domain (e.g. jepa-fx, hyperguild); a hall is one of a closed
vocabulary of memory types (facts, decisions, failures, hypotheses,
sources). Notes route to brain/wiki/<wing>/<hall>/<slug>.md with
wing/hall/created_at YAML frontmatter, making the directory a valid
Obsidian vault.
Changes:
- new package ingestion/internal/brain (NotePath, ValidHalls, Sanitise,
BuildWingIndex, BuildAllWingIndexes)
- api.WriteNote refactored to WriteNoteOptions; wing+hall routes to
brain/wiki/, otherwise falls back to brain/knowledge/ (legacy)
- search.Query → QueryOptions with optional Wing/Hall filtering; Result
carries wing/hall extracted from frontmatter or path segments
- MCP tools brain_write and brain_query gain optional wing/hall params
(hall enum-validated); new brain_index tool regenerates _index.md MOC
- POST /index REST endpoint mirrors brain_index
- brain_write auto-rebuilds the wing's _index.md after a wing+hall write
- scripts/migrate-brain-halls.sh migrates flat brain/wiki/{concepts,entities}/
into the new layout (dry-run by default, --commit applies)
All existing tests pass; new tests cover wing/hall write routing, scope
filtering, invalid hall rejection, _index.md generation, and migration
script paths.
Closes hyperguild#1.
Adds two new LLM-backed MCP tools to the ingestion service:
- brain_answer(query): BM25 retrieval + LLM synthesis → answer + sources
- brain_classify(text): classifies doc into type/title/tags via LLM
Adds llm.Router for primary→fallback routing (berget.ai → iguana).
Wired via BRAIN_LLM_PRIMARY_URL/BRAIN_LLM_FALLBACK_URL env vars;
no-op when unset so existing deployments are unaffected.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
mathias
