Compare commits

...
Author SHA1 Message Date
mathiasandClaude Opus 4.8 b9d03316fd fix(ingestion): migrate mcp-chassis import to git.d-ma.be path (#74)
CI / Lint / Test / Vet (push) Successful in 13s
CI / Mirror to GitHub (push) Has been skipped
Unblocks CD: the build's go mod download failed because mcp-chassis was
imported via the old gitea.d-ma.be module path, which the renamed server no
longer serves a matching go-import meta tag for. Point at the renamed module
git.d-ma.be/mathias/mcp-chassis v0.2.0 (mcp-chassis 69d389d) + go mod tidy.
Fresh download now resolves; ingestion builds + tests green.

This restores the normal CI path so the routing pass-rate fix (da9bdc4,
#73) builds and deploys through gitea Actions. Refs #74, #73, #35.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-30 21:39:41 +02:00
mathiasandClaude Opus 4.8 da9bdc4cbb fix(routing): repair pass-rate instrumentation (3 bugs) — #73, #35
CI / Lint / Test / Vet (push) Successful in 13s
CI / Mirror to GitHub (push) Successful in 4s
The #35 data gate could never fill: a real review call routed cleanly to
qwen36 but /pass-rate stayed total:0 under every key. Root cause was three
independent defects in the session_log path, each alone fatal:

- A: a successful routed call logged final_status "skip", never "pass".
  /pass-rate computes pass/(pass+fail) and skips count as neither, so the
  >=0.90 gate was mathematically unreachable. Success now logs "pass".
- B: every record was written under skill "_routing", so /pass-rate?skill=
  review|debug (what #35 measures) always read zero. Now uses the real
  e.Skill; routing decisions stay groupable via session_id "_routing".
- C: the session_log POST to the bearer-gated ingestion /mcp carried no
  Authorization header → silent 401, swallowed by best-effort logging
  (the documented mcpclient-empty-token-silent-401 footgun). Logger now
  takes a token (BRAIN_MCP_TOKEN) and sets the bearer when non-empty.

Tests rewritten to assert correct behavior (they had encoded the bugs:
"skip" on success, "_routing" skill). New test covers the auth header and
the empty-token path.

Infra (BRAIN_MCP_TOKEN ExternalSecret + env on the routing deployment) and
redeploy follow separately. Refs #73, #35.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-30 08:36:18 +02:00
mathiasandClaude Opus 4.8 dcb9ff4a56 docs(runbook): how to exercise review/debug traffic for the pass-rate gate
CI / Lint / Test / Vet (push) Successful in 20s
CI / Mirror to GitHub (push) Successful in 3s
The #35 data gate stays at zero because pass-rate only accrues from review/
debug calls *through the routing pod* — Crush, cloud chat, and the local
.skills all bypass it. Document the connect → route → flywheel steps, the
endpoints (koala:30310/mcp + routing-mcp.d-ma.be), the cold-start behavior
(nil pass-rate routes cloud until passes accrue past the 0.90 floor, then
local qwen36 activates), the 50-invocation / 2026-07-10 target, and the
Berget fallback. Refs #35.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 23:41:07 +02:00
mathias 0454527b83 Merge pull request 'feat: brain_pending + brain_promote — close the raw→wiki curation loop (#38)' (#70) from feat/brain-pending-promote into main
CI / Lint / Test / Vet (push) Successful in 13s
CI / Mirror to GitHub (push) Successful in 3s
2026-06-26 14:49:27 +00:00
mathiasandClaude Opus 4.8 ef11864121 feat(mcp): brain_pending + brain_promote tools + REST routes (#38)
CI / Lint / Test / Vet (pull_request) Successful in 13s
CI / Mirror to GitHub (pull_request) Has been skipped
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>
2026-06-26 16:34:15 +02:00
mathiasandClaude Opus 4.8 14b04a25cb feat(brain): ListPending + PromoteNote — raw→wiki curation primitives (#38)
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>
2026-06-26 16:34:15 +02:00
mathias 66a9b8e725 Merge pull request 'docs(close-session): refresh classification gate post-#67 (#62)' (#69) from fix/close-session-classification-refresh into main
CI / Lint / Test / Vet (push) Successful in 12s
CI / Mirror to GitHub (push) Successful in 3s
2026-06-26 14:28:48 +00:00
mathiasandClaude Opus 4.8 9f8fb9c138 docs(close-session): refresh classification gate post-#67 (#62)
CI / Lint / Test / Vet (pull_request) Successful in 13s
CI / Mirror to GitHub (pull_request) Has been skipped
The capture-routing in 723dab5 (#62 minimal) carried pre-#67 prose: it
warned "no populated classification.yaml yet" and steered repos_touched
away from brain/ai-sessions as if they'd escalate to confidential. #67
tagged the homelab repos internal, so that guidance is now wrong and
over-restrictive — listing the central repos is fine, and the summary's
fixed ai-sessions target no longer escalates the call. Point at
classification.yaml as the source of truth; reserve the refusal warning
for client-*/untagged material. Completes #62's veneer.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-26 16:25:57 +02:00
mathias d39a18dd69 Merge pull request 'fix: wire ai-sessions SummaryWriter into the capture relay (#66)' (#68) from fix/capture-summarywriter into main
CI / Lint / Test / Vet (push) Successful in 12s
CI / Mirror to GitHub (push) Successful in 4s
2026-06-23 15:22:42 +00:00
mathias 723dab51ae feat(close-session): route closeout through the capture tool (#62 minimal)
CI / Lint / Test / Vet (push) Successful in 13s
CI / Mirror to GitHub (push) Successful in 3s
Collapse Phases 4 (summary commit) + 5 (brain note) into a single
capture-driven Phase 4. The skill now assembles one brain:capture payload
(insights + tickets + summary + context) and dry-runs-then-executes;
capture owns the brain/gitea/ai-sessions writes, the I1 gate, the I5
audit record, and the supersession/read-after-write discipline server-side.

Adds the classification-gate lesson learned dogfooding capture this
session: effective classification = strictest across ALL targets
(wings, ticket repos, repos_touched); untagged repos (brain, ai-sessions)
fail-safe to confidential and get refused via claude.ai, so repos_touched
must stay to internal-default repos. Legacy inline path kept only as the
capture-unreachable fallback. Staleness prose removed (server owns it now).

Partial of #62; harness-token + harvest-adapter + fidelity-supersession
work remain.
2026-06-23 06:27:57 +00:00
17 changed files with 628 additions and 86 deletions
+1 -1
View File
@@ -53,7 +53,7 @@ func main() {
router := &routing.Router{
Fetcher: routing.NewFetcher(cfg.BrainURL, "7d", time.Duration(cfg.PassRateTTLSeconds)*time.Second),
Logger: routing.NewLogger(cfg.BrainURL),
Logger: routing.NewLogger(cfg.BrainURL, cfg.BrainMCPToken),
Policy: routing.Policy{Floor: cfg.RouteLocalFloor, Ceil: cfg.RouteLocalCeil},
FastModel: cfg.FastModel,
ThinkingModel: cfg.ThinkingModel,
@@ -0,0 +1,95 @@
# Runbook: exercising review/debug traffic to fill the pass-rate dataset
**Why this exists:** the routing pod's local-vs-cloud decision is gated on a
pass-rate history that only accrues from real `review`/`debug` invocations
**through the pod**. Until the dataset has data, the fast (local) path never
activates and the core hypothesis (hyperguild #35) can't be validated. This
runbook is how you spin that flywheel.
## The one trap
Pass-rate accrues **only** when a skill tool is called via the routing pod's MCP
endpoint. These look like they should count but **do not**:
- **Crush** — talks to LiteLLM directly, bypasses the pod. No log.
- **claude.ai web / Claude Desktop without the connector** — no log.
- **Running the local `code-review` / `debug` skills** (`~/dev/.skills`) inline in
a Claude Code session — those are local skills, not the pod's MCP tools. No log.
Only a `tools/call` to the routing pod records a pass/fail.
## Endpoints
| Purpose | URL | Auth |
|---------|-----|------|
| Routing MCP (local, Tailscale) | `http://koala:30310/mcp` | Bearer `ROUTING_MCP_TOKEN` |
| Routing MCP (remote) | `https://routing-mcp.d-ma.be/mcp` | OAuth via `auth.d-ma.be` (audience `claude-ai`) |
| Pass-rate readout | `http://koala:30330/pass-rate?skill=<name>` | none (read-only) |
Tools advertised: **`review`**, **`debug`** (the two the #35 gate measures),
plus `session_log`, `retrospective`, `trainer`.
## Step 1 — connect the routing pod as an MCP server
**Local** (needs the bearer token; keep it out of argv via 1Password):
```bash
op run --env-file ~/.op-env -- \
claude mcp add routing --transport http http://koala:30310/mcp \
--header "Authorization: Bearer $ROUTING_MCP_TOKEN"
```
**Remote** (claude.ai / Claude Desktop): add a custom connector pointing at
`https://routing-mcp.d-ma.be/mcp`; it completes OAuth against `auth.d-ma.be`,
no static token.
Verify: a `tools/list` should return `review`, `debug`, `session_log`,
`retrospective`, `trainer`.
## Step 2 — route real work through it
In normal sessions, invoke the pod's tools instead of reviewing/debugging inline:
- *"Use the **routing** `review` tool on this diff."*
- *"**debug** this failure through the routing pod."*
Each call logs an outcome to ingestion → `/pass-rate` ticks up.
## Step 3 — the flywheel (cold-start behavior is the gate)
The router's cold start **is** the data gate — same loop:
1. With pass-rate `nil` (cold), the policy routes to the **thinking / cloud**
model — it won't trust the local fast model without history.
2. As `review` / `debug` accumulate **passes**, pass-rate climbs past the floor
(`HYPERGUILD_ROUTE_LOCAL_FLOOR=0.90`) and the router starts sending those
skills to the **local fast tier**.
3. So early calls "pay in" on cloud to build the record; then the local tier
activates automatically. Exercising the traffic is what spins it up.
The fast tier is now **`koala/qwen36-35b-a3b`** (Qwen3.6-35B-A3B MTP), promoted
2026-06-29 after clearing the tool-call gate (infra `c66a195`,
`HYPERGUILD_FAST_MODEL`).
## Target & verification
- **50 logged invocations** across `review` + `debug` within the 14-day window
(opened 2026-06-26, **kill-date 2026-07-10**) → ~45 calls/day.
- Check progress anytime:
```bash
curl -s "http://koala:30330/pass-rate?skill=review"
curl -s "http://koala:30330/pass-rate?skill=debug"
```
- If ~45/day isn't realistic alongside Crush, that is **not** a failure — per
#35 deliverable #1 it's the signal hyperguild isn't on the work critical path,
and the pre-decided **Berget fallback** (`gpt-oss-120b` / `qwen3-32b`) carries
the fast tier instead.
## Refs
- hyperguild #35 — the validation issue (data gate = deliverable #1)
- `docs/multi-model-routing.md` — routing policy
- brain: `wiki/homelab/hypotheses/qwen36-35b-a3b-fast-model-experiment-2026-05-28.md`
- infra `c66a195` — qwen36 promotion; `models.yml` / `llama-swap-configmap.yaml`
+3 -1
View File
@@ -13,7 +13,7 @@ import (
"strings"
"time"
chassisauth "gitea.d-ma.be/mathias/mcp-chassis/auth"
chassisauth "git.d-ma.be/mathias/mcp-chassis/auth"
"github.com/mathiasbq/hyperguild/ingestion/internal/api"
"github.com/mathiasbq/hyperguild/ingestion/internal/audit"
@@ -369,6 +369,8 @@ func main() {
mux.HandleFunc("POST /ingest-path", h.IngestPath)
mux.HandleFunc("POST /ingest-raw", h.IngestRaw)
mux.HandleFunc("POST /backfill-refs", h.BackfillRefs)
mux.HandleFunc("GET /pending", h.Pending)
mux.HandleFunc("POST /promote", h.Promote)
mux.HandleFunc("POST /backfill-embeddings", h.BackfillEmbeddings)
mux.HandleFunc("GET /pass-rate", h.PassRate)
jwtValidator, err := chassisauth.NewJWTValidator(ctx, os.Getenv("DEX_ISSUER_URL"), os.Getenv("MCP_AUDIENCE"))
+8 -5
View File
@@ -2,19 +2,22 @@ module github.com/mathiasbq/hyperguild/ingestion
go 1.26.1
require github.com/stretchr/testify v1.11.1
require (
github.com/lestrrat-go/jwx/v2 v2.1.6
github.com/stretchr/testify v1.11.1
github.com/kr/text v0.2.0 // indirect
github.com/lestrrat-go/jwx/v2 v2.1.6 // indirect
github.com/rogpeppe/go-internal v1.15.0 // indirect
)
require (
gitea.d-ma.be/mathias/mcp-chassis v0.1.0 // indirect
git.d-ma.be/mathias/mcp-chassis v0.2.0
github.com/davecgh/go-spew v1.1.1 // indirect
github.com/decred/dcrd/dcrec/secp256k1/v4 v4.4.0 // indirect
github.com/goccy/go-json v0.10.3 // indirect
github.com/jackc/pgpassfile v1.0.0 // indirect
github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 // indirect
github.com/jackc/pgx/v5 v5.9.2 // indirect
github.com/jackc/pgx/v5 v5.9.2
github.com/jackc/puddle/v2 v2.2.2 // indirect
github.com/lestrrat-go/blackmagic v1.0.3 // indirect
github.com/lestrrat-go/httpcc v1.0.1 // indirect
@@ -27,5 +30,5 @@ require (
golang.org/x/sync v0.17.0 // indirect
golang.org/x/sys v0.31.0 // indirect
golang.org/x/text v0.29.0 // indirect
gopkg.in/yaml.v3 v3.0.1 // indirect
gopkg.in/yaml.v3 v3.0.1
)
+10 -3
View File
@@ -1,5 +1,6 @@
gitea.d-ma.be/mathias/mcp-chassis v0.1.0 h1:8RXO34+n7Vu8HnUMagars6fc4oemqRpMu7MVtjaj4qY=
gitea.d-ma.be/mathias/mcp-chassis v0.1.0/go.mod h1:ajbLlwr2L7FAN3TBU39KucZkKJM02wTbKbDKDEW2YvE=
git.d-ma.be/mathias/mcp-chassis v0.2.0 h1:6fLmb7xqRa2nNVWsHaUbbfbArgDXJw/gDhb09clBIjo=
git.d-ma.be/mathias/mcp-chassis v0.2.0/go.mod h1:Ks7EK2UnGAN0H3rJjKUxUagX8/ZBdtLrOlcUbv0RwH8=
github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
@@ -15,6 +16,10 @@ github.com/jackc/pgx/v5 v5.9.2 h1:3ZhOzMWnR4yJ+RW1XImIPsD1aNSz4T4fyP7zlQb56hw=
github.com/jackc/pgx/v5 v5.9.2/go.mod h1:mal1tBGAFfLHvZzaYh77YS/eC6IX9OWbRV1QIIM0Jn4=
github.com/jackc/puddle/v2 v2.2.2 h1:PR8nw+E/1w0GLuRFSmiioY6UooMp6KJv0/61nB7icHo=
github.com/jackc/puddle/v2 v2.2.2/go.mod h1:vriiEXHvEE654aYKXXjOvZM39qJ0q+azkZFrfEOc3H4=
github.com/kr/pretty v0.3.0 h1:WgNl7dwNpEZ6jJ9k1snq4pZsg7DOEN8hP9Xw0Tsjwk0=
github.com/kr/pretty v0.3.0/go.mod h1:640gp4NfQd8pI5XOwp5fnNeVWj67G7CFk/SaSQn7NBk=
github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
github.com/lestrrat-go/blackmagic v1.0.3 h1:94HXkVLxkZO9vJI/w2u1T0DAoprShFd13xtnSINtDWs=
github.com/lestrrat-go/blackmagic v1.0.3/go.mod h1:6AWFyKNNj0zEXQYfTMPfZrAXUWUfTIZ5ECEUEJaijtw=
github.com/lestrrat-go/httpcc v1.0.1 h1:ydWCStUeJLkpYyjLDHihupbn2tYmZ7m22BGkcvZZrIE=
@@ -29,6 +34,8 @@ github.com/lestrrat-go/option v1.0.1 h1:oAzP2fvZGQKWkvHa1/SAcFolBEca1oN+mQ7eooNB
github.com/lestrrat-go/option v1.0.1/go.mod h1:5ZHFbivi4xwXxhxY9XHDe2FHo6/Z7WWmtT7T5nBBp3I=
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
github.com/rogpeppe/go-internal v1.15.0 h1:D0RCU5rMAp+SpgkiNdrjfJ+LX4J1M32V2NeCY7EJ6hc=
github.com/rogpeppe/go-internal v1.15.0/go.mod h1:DrUVZyrJU+txYW5/1kwtXQSMFio52ZOxX7yM1VHvnxs=
github.com/segmentio/asm v1.2.0 h1:9BQrFxC+YOHJlTlHGkTrFWf59nbL3XnCoFLTwDCI7ys=
github.com/segmentio/asm v1.2.0/go.mod h1:BqMnlJP91P8d+4ibuonYZw9mfnzI9HfxselHZr5aAcs=
github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
@@ -46,9 +53,9 @@ golang.org/x/sys v0.31.0 h1:ioabZlmFYtWhL+TRYpcnNlLwhyxaM9kWTDEmfnprqik=
golang.org/x/sys v0.31.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k=
golang.org/x/text v0.29.0 h1:1neNs90w9YzJ9BocxfsQNHKuAT4pkghyXc4nhZ6sJvk=
golang.org/x/text v0.29.0/go.mod h1:7MhJOA9CD2qZyOKYazxdYMF85OwPdEr9jTtBpO7ydH4=
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c h1:Hei/4ADfdWqJk1ZMxUNpqntNwaWcugrBjAiHlqqRiVk=
gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c/go.mod h1:JHkPIbrfpd72SG/EVd6muEfDQjcINNoR0C8j2r3qZ4Q=
gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
+34
View File
@@ -483,6 +483,40 @@ func (h *Handler) BackfillRefs(w http.ResponseWriter, r *http.Request) {
writeJSON(w, map[string]int{"updated": n})
}
// Pending handles GET /pending — list raw/ notes awaiting promotion.
func (h *Handler) Pending(w http.ResponseWriter, _ *http.Request) {
pending, err := ListPending(h.brainDir)
if err != nil {
h.logger.Error("pending failed", "err", err)
writeError(w, http.StatusInternalServerError, "pending error")
return
}
writeJSON(w, map[string]any{"pending": pending})
}
type promoteRequest struct {
Filename string `json:"filename"`
Wing string `json:"wing"`
Hall string `json:"hall"`
Slug string `json:"slug,omitempty"`
}
// Promote handles POST /promote — move a raw/ note into the wiki. A bad
// hall / collision / missing source is a 400 (caller error), not a 500.
func (h *Handler) Promote(w http.ResponseWriter, r *http.Request) {
var req promoteRequest
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
writeError(w, http.StatusBadRequest, "invalid JSON")
return
}
rel, err := PromoteNote(h.brainDir, PromoteOptions(req))
if err != nil {
writeError(w, http.StatusBadRequest, err.Error())
return
}
writeJSON(w, map[string]string{"path": rel})
}
func writeJSON(w http.ResponseWriter, v any) {
w.Header().Set("Content-Type", "application/json")
json.NewEncoder(w).Encode(v) //nolint:errcheck
+156
View File
@@ -0,0 +1,156 @@
package api
import (
"fmt"
"os"
"path/filepath"
"regexp"
"sort"
"strings"
"time"
"github.com/mathiasbq/hyperguild/ingestion/internal/brain"
)
// PendingNote describes a raw/ note awaiting human promotion to the wiki.
type PendingNote struct {
Filename string `json:"filename"`
CreatedAt string `json:"created_at"`
SizeBytes int64 `json:"size_bytes"`
Excerpt string `json:"excerpt"`
}
// datePrefix matches a leading YYYY-MM-DD- on a raw filename, stripped when
// deriving the default promoted slug.
var datePrefix = regexp.MustCompile(`^\d{4}-\d{2}-\d{2}-`)
// ListPending returns the notes in brain/raw/ awaiting review, oldest-first
// (natural review order). An absent raw/ dir yields an empty slice, not an
// error. Only .md files are listed; tunnel-candidate files are skipped.
func ListPending(brainDir string) ([]PendingNote, error) {
dir := filepath.Join(brainDir, "raw")
entries, err := os.ReadDir(dir)
if err != nil {
if os.IsNotExist(err) {
return []PendingNote{}, nil
}
return nil, fmt.Errorf("read raw dir: %w", err)
}
out := make([]PendingNote, 0, len(entries))
for _, e := range entries {
if e.IsDir() || !strings.HasSuffix(e.Name(), ".md") || strings.HasPrefix(e.Name(), "tunnel-candidates-") {
continue
}
info, statErr := e.Info()
if statErr != nil {
continue
}
raw, readErr := os.ReadFile(filepath.Join(dir, e.Name()))
if readErr != nil {
continue
}
fm, body := parseFrontmatter(string(raw))
created := fm.get("created_at")
if created == "" {
created = info.ModTime().UTC().Format(time.RFC3339)
}
out = append(out, PendingNote{
Filename: e.Name(),
CreatedAt: created,
SizeBytes: info.Size(),
Excerpt: excerpt(body, 200),
})
}
sort.SliceStable(out, func(i, j int) bool { return out[i].CreatedAt < out[j].CreatedAt })
return out, nil
}
// PromoteOptions identifies a raw note to promote and its wiki destination.
type PromoteOptions struct {
Filename string // basename in brain/raw/
Wing string
Hall string
Slug string // optional; defaults to Filename minus date prefix + .md
}
// PromoteNote moves a note from brain/raw/ into the structured wiki: it
// rewrites frontmatter (sets wing/hall/promoted_at, preserves created_at and
// any custom fields), writes to brain/wiki/<wing>/<hall>/<slug>.md, deletes
// the source, then rebuilds the wing index and runs auto-tunnel detection.
//
// It is atomic from the caller's view: validation (hall, wing, slug,
// collision) happens before any filesystem change, and the source is deleted
// only after the destination write succeeds (write-then-delete, never move).
// Returns the promoted note's path relative to brainDir.
func PromoteNote(brainDir string, opts PromoteOptions) (string, error) {
// Validate filename (basename only — no traversal) before touching fs.
base := filepath.Base(opts.Filename)
if base != opts.Filename || base == "." || base == ".." || strings.ContainsAny(opts.Filename, `/\`) {
return "", fmt.Errorf("invalid filename %q", opts.Filename)
}
slug := opts.Slug
if slug == "" {
slug = datePrefix.ReplaceAllString(strings.TrimSuffix(base, ".md"), "")
}
// NotePath validates hall + wing + slug; do this before reading anything.
dest, err := brain.NotePath(brainDir, opts.Wing, opts.Hall, slug)
if err != nil {
return "", err
}
src := filepath.Join(brainDir, "raw", base)
raw, err := os.ReadFile(src)
if err != nil {
if os.IsNotExist(err) {
return "", fmt.Errorf("pending note %q does not exist in raw/", base)
}
return "", fmt.Errorf("read source: %w", err)
}
// Collision: never silently overwrite an existing promoted note.
if _, statErr := os.Stat(dest); statErr == nil {
rel, _ := filepath.Rel(brainDir, dest)
return "", fmt.Errorf("target %s already exists; choose a different slug", filepath.ToSlash(rel))
}
fm, body := parseFrontmatter(string(raw))
now := time.Now().UTC().Format(time.RFC3339)
fm.set("wing", brain.Sanitise(opts.Wing))
fm.set("hall", opts.Hall)
if fm.get("created_at") == "" {
fm.set("created_at", now)
}
fm.set("promoted_at", now)
if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil {
return "", fmt.Errorf("create wing dir: %w", err)
}
// Write-then-delete: the source survives any write failure.
if err := os.WriteFile(dest, []byte(fm.render()+body), 0o644); err != nil {
return "", fmt.Errorf("write promoted note: %w", err)
}
if err := os.Remove(src); err != nil {
return "", fmt.Errorf("promoted note written but source removal failed: %w", err)
}
rel, _ := filepath.Rel(brainDir, dest)
relSlash := filepath.ToSlash(rel)
// Best-effort wiki upkeep — the note is already promoted.
_ = brain.BuildWingIndex(brainDir, opts.Wing)
_ = brain.AutoTunnel(brainDir, relSlash, body)
return relSlash, nil
}
// excerpt returns the first n runes of s, trimmed, single-spaced.
func excerpt(s string, n int) string {
s = strings.TrimSpace(s)
r := []rune(s)
if len(r) > n {
r = r[:n]
}
return strings.TrimSpace(string(r))
}
+121
View File
@@ -0,0 +1,121 @@
package api
import (
"os"
"path/filepath"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func writeRaw(t *testing.T, brainDir, name, content string) {
t.Helper()
dir := filepath.Join(brainDir, "raw")
require.NoError(t, os.MkdirAll(dir, 0o755))
require.NoError(t, os.WriteFile(filepath.Join(dir, name), []byte(content), 0o644))
}
func TestListPendingEmptyWhenAbsent(t *testing.T) {
got, err := ListPending(t.TempDir())
require.NoError(t, err, "absent raw/ is not an error")
assert.Empty(t, got)
}
func TestListPendingReturnsOldestFirstWithExcerpt(t *testing.T) {
dir := t.TempDir()
writeRaw(t, dir, "2026-06-02-newer.md", "---\ncreated_at: 2026-06-02T00:00:00Z\n---\nNewer body here.\n")
writeRaw(t, dir, "2026-06-01-older.md", "---\ncreated_at: 2026-06-01T00:00:00Z\n---\nOlder body content.\n")
// non-md ignored
writeRaw(t, dir, "notes.txt", "ignore me")
got, err := ListPending(dir)
require.NoError(t, err)
require.Len(t, got, 2)
assert.Equal(t, "2026-06-01-older.md", got[0].Filename, "oldest first")
assert.Equal(t, "2026-06-02-newer.md", got[1].Filename)
assert.Contains(t, got[0].Excerpt, "Older body content")
assert.NotContains(t, got[0].Excerpt, "---", "excerpt is body, not frontmatter")
assert.Positive(t, got[0].SizeBytes)
}
func TestPromoteHappyPath(t *testing.T) {
dir := t.TempDir()
writeRaw(t, dir, "2026-06-01-lejpa-decision.md",
"---\ncreated_at: 2026-06-01T09:00:00Z\ncustom_field: keep-me\n---\n# LeJEPA\n\nbody.\n")
rel, err := PromoteNote(dir, PromoteOptions{
Filename: "2026-06-01-lejpa-decision.md", Wing: "jepa-fx", Hall: "decisions",
})
require.NoError(t, err)
assert.Equal(t, "wiki/jepa-fx/decisions/lejpa-decision.md", rel, "slug defaults to filename minus date prefix")
// Source deleted.
_, statErr := os.Stat(filepath.Join(dir, "raw", "2026-06-01-lejpa-decision.md"))
assert.True(t, os.IsNotExist(statErr), "source removed after promote")
got, err := os.ReadFile(filepath.Join(dir, filepath.FromSlash(rel)))
require.NoError(t, err)
s := string(got)
assert.Contains(t, s, "wing: jepa-fx")
assert.Contains(t, s, "hall: decisions")
assert.Contains(t, s, "created_at: 2026-06-01T09:00:00Z", "original created_at preserved")
assert.Contains(t, s, "promoted_at:")
assert.Contains(t, s, "custom_field: keep-me", "custom frontmatter preserved")
assert.Contains(t, s, "# LeJEPA")
}
func TestPromoteExplicitSlug(t *testing.T) {
dir := t.TempDir()
writeRaw(t, dir, "2026-06-01-x.md", "body\n")
rel, err := PromoteNote(dir, PromoteOptions{Filename: "2026-06-01-x.md", Wing: "a", Hall: "facts", Slug: "custom-slug"})
require.NoError(t, err)
assert.Equal(t, "wiki/a/facts/custom-slug.md", rel)
}
func TestPromoteInvalidHallErrorsBeforeTouchingFS(t *testing.T) {
dir := t.TempDir()
writeRaw(t, dir, "2026-06-01-x.md", "body\n")
_, err := PromoteNote(dir, PromoteOptions{Filename: "2026-06-01-x.md", Wing: "a", Hall: "garbage"})
require.Error(t, err)
// Source untouched.
_, statErr := os.Stat(filepath.Join(dir, "raw", "2026-06-01-x.md"))
assert.NoError(t, statErr, "invalid hall must not delete or move the source")
}
func TestPromoteMissingSourceErrors(t *testing.T) {
_, err := PromoteNote(t.TempDir(), PromoteOptions{Filename: "ghost.md", Wing: "a", Hall: "facts"})
require.Error(t, err)
}
func TestPromoteSlugCollisionNoOverwrite(t *testing.T) {
dir := t.TempDir()
// Pre-existing target.
dest := filepath.Join(dir, "wiki", "a", "facts", "x.md")
require.NoError(t, os.MkdirAll(filepath.Dir(dest), 0o755))
require.NoError(t, os.WriteFile(dest, []byte("EXISTING\n"), 0o644))
writeRaw(t, dir, "2026-06-01-x.md", "NEW\n")
_, err := PromoteNote(dir, PromoteOptions{Filename: "2026-06-01-x.md", Wing: "a", Hall: "facts"})
require.Error(t, err, "collision must error, not overwrite")
got, _ := os.ReadFile(dest)
assert.Equal(t, "EXISTING\n", string(got), "target not overwritten")
_, statErr := os.Stat(filepath.Join(dir, "raw", "2026-06-01-x.md"))
assert.NoError(t, statErr, "source preserved on collision (atomic: no delete without write)")
}
func TestPromoteRejectsTraversalFilename(t *testing.T) {
_, err := PromoteNote(t.TempDir(), PromoteOptions{Filename: "../escape.md", Wing: "a", Hall: "facts"})
require.Error(t, err)
}
func TestPromoteRebuildsWingIndex(t *testing.T) {
dir := t.TempDir()
writeRaw(t, dir, "2026-06-01-x.md", "---\ntitle: X Note\n---\nbody\n")
_, err := PromoteNote(dir, PromoteOptions{Filename: "2026-06-01-x.md", Wing: "a", Hall: "facts"})
require.NoError(t, err)
idx, err := os.ReadFile(filepath.Join(dir, "wiki", "a", "_index.md"))
require.NoError(t, err, "wing _index regenerated")
assert.Contains(t, string(idx), "x", "promoted note appears in the index")
}
+50
View File
@@ -9,6 +9,7 @@ import (
"strings"
"time"
"github.com/mathiasbq/hyperguild/ingestion/internal/api"
"github.com/mathiasbq/hyperguild/ingestion/internal/brain"
"github.com/mathiasbq/hyperguild/ingestion/internal/capture"
"github.com/mathiasbq/hyperguild/ingestion/internal/extract"
@@ -81,6 +82,21 @@ func (s *Server) tools() []map[string]any {
"path": str("brain-relative path to the note; equivalent to id"),
}),
},
{
"name": "brain_pending",
"description": "List notes in brain/raw/ awaiting human promotion to the wiki, oldest-first. Returns filename, created_at, size_bytes, excerpt. The human-review queue complement to brain_promote.",
"inputSchema": schema([]string{}, map[string]any{}),
},
{
"name": "brain_promote",
"description": "Promote a brain/raw/ note into brain/wiki/<wing>/<hall>/: rewrites frontmatter (sets wing/hall/promoted_at, preserves created_at + custom fields), deletes the source, rebuilds the wing index, runs auto-tunnel. Errors (without touching the fs) on invalid hall or a slug collision. Returns {path}.",
"inputSchema": schema([]string{"filename", "wing", "hall"}, map[string]any{
"filename": str("basename in brain/raw/, e.g. 2026-06-01-lejpa-decision.md"),
"wing": str("target wing, e.g. jepa-fx"),
"hall": enum("target hall", halls...),
"slug": str("optional target slug; defaults to filename minus date prefix"),
}),
},
{
"name": "brain_tunnel",
"description": "Create an explicit bidirectional [[wikilink]] between two notes in different wings. Idempotent.",
@@ -321,6 +337,40 @@ func (s *Server) brainGet(ctx context.Context, args json.RawMessage) (json.RawMe
})
}
// brainPending lists the raw/ review queue (oldest-first).
func (s *Server) brainPending(_ context.Context, _ json.RawMessage) (json.RawMessage, error) {
pending, err := api.ListPending(s.brainDir)
if err != nil {
return nil, err
}
return json.Marshal(map[string]any{"pending": pending})
}
type brainPromoteArgs struct {
Filename string `json:"filename"`
Wing string `json:"wing"`
Hall string `json:"hall"`
Slug string `json:"slug,omitempty"`
}
// brainPromote moves a raw/ note into the structured wiki (frontmatter
// rewrite + index + auto-tunnel, all owned by api.PromoteNote) and then
// re-indexes it into the graph. The human-facing complement to brain_write.
func (s *Server) brainPromote(ctx context.Context, args json.RawMessage) (json.RawMessage, error) {
var a brainPromoteArgs
if err := json.Unmarshal(args, &a); err != nil {
return nil, fmt.Errorf("parse args: %w", err)
}
relPath, err := api.PromoteNote(s.brainDir, api.PromoteOptions{
Filename: a.Filename, Wing: a.Wing, Hall: a.Hall, Slug: a.Slug,
})
if err != nil {
return nil, err
}
s.indexInGraph(ctx, "brain_promote", relPath)
return json.Marshal(map[string]string{"path": relPath})
}
// indexInGraph is a best-effort wrapper around graphsync.IndexDoc that
// logs failures but never propagates them — the underlying write/ingest
// has already succeeded and the graph is an augmentation, not a
+58
View File
@@ -332,3 +332,61 @@ func TestSessionLogRequiresSessionID(t *testing.T) {
resp := toolCall(t, srv, "session_log", map[string]any{"skill": "tdd"})
require.NotNil(t, resp["error"])
}
func TestBrainPendingListsRaw(t *testing.T) {
brainDir := t.TempDir()
raw := filepath.Join(brainDir, "raw")
require.NoError(t, os.MkdirAll(raw, 0o755))
require.NoError(t, os.WriteFile(filepath.Join(raw, "2026-06-01-x.md"),
[]byte("---\ncreated_at: 2026-06-01T00:00:00Z\n---\npending body\n"), 0o644))
srv := mcp.NewServer(brainDir, nil, nil, nil)
resp := toolCall(t, srv, "brain_pending", map[string]any{})
require.Nil(t, resp["error"])
text := resp["result"].(map[string]any)["content"].([]any)[0].(map[string]any)["text"].(string)
assert.Contains(t, text, "2026-06-01-x.md")
assert.Contains(t, text, "pending body")
}
func TestBrainPendingEmpty(t *testing.T) {
srv := mcp.NewServer(t.TempDir(), nil, nil, nil)
resp := toolCall(t, srv, "brain_pending", map[string]any{})
require.Nil(t, resp["error"])
text := resp["result"].(map[string]any)["content"].([]any)[0].(map[string]any)["text"].(string)
assert.Contains(t, text, `"pending":[]`)
}
func TestBrainPromoteMovesToWiki(t *testing.T) {
brainDir := t.TempDir()
raw := filepath.Join(brainDir, "raw")
require.NoError(t, os.MkdirAll(raw, 0o755))
require.NoError(t, os.WriteFile(filepath.Join(raw, "2026-06-01-decision.md"),
[]byte("---\ncreated_at: 2026-06-01T00:00:00Z\n---\n# D\n\nbody\n"), 0o644))
srv := mcp.NewServer(brainDir, nil, nil, nil)
resp := toolCall(t, srv, "brain_promote", map[string]any{
"filename": "2026-06-01-decision.md", "wing": "jepa-fx", "hall": "decisions",
})
require.Nil(t, resp["error"], "got: %v", resp["error"])
text := resp["result"].(map[string]any)["content"].([]any)[0].(map[string]any)["text"].(string)
assert.Contains(t, text, "wiki/jepa-fx/decisions/decision.md")
_, err := os.Stat(filepath.Join(brainDir, "wiki/jepa-fx/decisions/decision.md"))
require.NoError(t, err)
_, srcErr := os.Stat(filepath.Join(raw, "2026-06-01-decision.md"))
assert.True(t, os.IsNotExist(srcErr), "source removed")
}
func TestBrainPromoteInvalidHallErrors(t *testing.T) {
brainDir := t.TempDir()
raw := filepath.Join(brainDir, "raw")
require.NoError(t, os.MkdirAll(raw, 0o755))
require.NoError(t, os.WriteFile(filepath.Join(raw, "x.md"), []byte("body\n"), 0o644))
srv := mcp.NewServer(brainDir, nil, nil, nil)
resp := toolCall(t, srv, "brain_promote", map[string]any{
"filename": "x.md", "wing": "a", "hall": "garbage",
})
require.NotNil(t, resp["error"])
_, srcErr := os.Stat(filepath.Join(raw, "x.md"))
assert.NoError(t, srcErr, "source untouched on validation error")
}
+8 -3
View File
@@ -1,8 +1,9 @@
// Package mcp implements an MCP HTTP handler for the ingestion service.
// Exposed tools: brain_query, brain_write, brain_update, brain_get,
// brain_index, brain_tunnel, brain_ingest, brain_ingest_raw,
// brain_answer, brain_classify, brain_graph, brain_context, session_log,
// and capture (the #55 relay tool, registered only when WithCapture is set).
// brain_pending, brain_promote, brain_index, brain_tunnel, brain_ingest,
// brain_ingest_raw, brain_answer, brain_classify, brain_graph,
// brain_context, session_log, and capture (the #55 relay tool, registered
// only when WithCapture is set).
package mcp
import (
@@ -263,6 +264,10 @@ func (s *Server) handleCall(ctx context.Context, name string, args json.RawMessa
return s.brainUpdate(ctx, args)
case "brain_get":
return s.brainGet(ctx, args)
case "brain_pending":
return s.brainPending(ctx, args)
case "brain_promote":
return s.brainPromote(ctx, args)
case "capture":
return s.brainCapture(ctx, args)
case "brain_index":
+1
View File
@@ -58,6 +58,7 @@ func TestServerToolsList(t *testing.T) {
}
assert.ElementsMatch(t, []string{
"brain_query", "brain_write", "brain_update", "brain_get",
"brain_pending", "brain_promote",
"brain_index", "brain_tunnel",
"brain_ingest_raw", "brain_ingest",
"brain_answer", "brain_classify", "brain_graph", "brain_context",
+2
View File
@@ -14,6 +14,7 @@ type RoutingConfig struct {
LiteLLMBaseURL string // LITELLM_BASE_URL, default https://llm-api.d-ma.be
LiteLLMAPIKey string // LITELLM_API_KEY
BrainURL string // BRAIN_URL, default http://ingestion.supervisor:3300
BrainMCPToken string // BRAIN_MCP_TOKEN, bearer for the auth-gated ingestion /mcp (session_log)
FastModel string // HYPERGUILD_FAST_MODEL, default koala/qwen35-9b-fast
ThinkingModel string // HYPERGUILD_THINKING_MODEL, default iguana/gemma4-26b
// RouteLocalFloor and RouteLocalCeil intentionally invert the usual
@@ -44,6 +45,7 @@ func LoadRouting() (RoutingConfig, error) {
LiteLLMBaseURL: envOr("LITELLM_BASE_URL", "https://llm-api.d-ma.be"),
LiteLLMAPIKey: os.Getenv("LITELLM_API_KEY"),
BrainURL: envOr("BRAIN_URL", "http://ingestion.supervisor:3300"),
BrainMCPToken: os.Getenv("BRAIN_MCP_TOKEN"),
FastModel: envOr("HYPERGUILD_FAST_MODEL", "koala/qwen35-9b-fast"),
ThinkingModel: envOr("HYPERGUILD_THINKING_MODEL", "iguana/gemma4-26b"),
}
+18 -5
View File
@@ -17,19 +17,24 @@ type LogEntry struct {
Message string // free-form, e.g. "model=qwen35, pass_rate=0.94"
ProjectRoot string
DurationMs int64
Failed bool // true → final_status: "fail"; false → "skip"
Failed bool // true → final_status: "fail"; false → "pass"
}
// Logger posts session_log entries to a brain MCP at BrainURL + /mcp.
type Logger struct {
BrainURL string
Token string // bearer for the (auth-gated) ingestion /mcp; empty = no header
HTTP *http.Client
}
// NewLogger creates a Logger with a 2-second HTTP timeout.
func NewLogger(brainURL string) *Logger {
// NewLogger creates a Logger with a 2-second HTTP timeout. token authenticates
// to the bearer-gated ingestion /mcp; an empty token sends no Authorization
// header (and silently 401s against a gated server — see brain
// mcpclient-empty-token-silent-401-envfrom-missing-key).
func NewLogger(brainURL, token string) *Logger {
return &Logger{
BrainURL: brainURL,
Token: token,
HTTP: &http.Client{Timeout: 2 * time.Second},
}
}
@@ -37,7 +42,10 @@ func NewLogger(brainURL string) *Logger {
// LogDecision posts a session_log MCP call. Errors are returned but the caller
// MUST NOT block real work on them — logging is best-effort.
func (l *Logger) LogDecision(ctx context.Context, e LogEntry) error {
status := "skip"
// A completed routed call is a pass (liveness); only an execution error is a
// fail. There is no "skip" for routing — the prior default-to-"skip" meant a
// successful call never counted toward pass_rate, so the gate was unreachable.
status := "pass"
if e.Failed {
status = "fail"
}
@@ -49,7 +57,9 @@ func (l *Logger) LogDecision(ctx context.Context, e LogEntry) error {
"name": "session_log",
"arguments": map[string]any{
"session_id": e.SessionID,
"skill": "_routing",
// The real skill, so /pass-rate?skill=review|debug sees these
// records; routing decisions stay groupable via session_id "_routing".
"skill": e.Skill,
"phase": "decide",
"final_status": status,
"message": fmt.Sprintf("%s: %s — %s", e.Skill, e.Decision, e.Message),
@@ -67,6 +77,9 @@ func (l *Logger) LogDecision(ctx context.Context, e LogEntry) error {
return fmt.Errorf("log: build request: %w", err)
}
req.Header.Set("Content-Type", "application/json")
if l.Token != "" {
req.Header.Set("Authorization", "Bearer "+l.Token)
}
resp, err := l.HTTP.Do(req)
if err != nil {
return fmt.Errorf("log: request: %w", err)
+31 -6
View File
@@ -15,36 +15,44 @@ import (
func TestLoggerLogDecision(t *testing.T) {
var captured map[string]any
var authHeader string
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
assert.Equal(t, http.MethodPost, r.Method)
assert.Equal(t, "/mcp", r.URL.Path)
authHeader = r.Header.Get("Authorization")
body, _ := io.ReadAll(r.Body)
require.NoError(t, json.Unmarshal(body, &captured))
_ = json.NewEncoder(w).Encode(map[string]any{"jsonrpc": "2.0", "id": 1, "result": map[string]any{"content": []map[string]any{{"type": "text", "text": "ok"}}}})
}))
defer srv.Close()
l := routing.NewLogger(srv.URL)
l := routing.NewLogger(srv.URL, "test-token")
err := l.LogDecision(context.Background(), routing.LogEntry{
SessionID: "sess-1",
Skill: "review",
Decision: "local",
Message: "model=qwen35, pass_rate=0.94",
Message: "model=qwen36, pass_rate=0.94",
ProjectRoot: "/home/x/proj",
DurationMs: 1234,
Failed: false,
})
require.NoError(t, err)
// Bug C fix: the POST authenticates to the bearer-gated ingestion /mcp.
assert.Equal(t, "Bearer test-token", authHeader)
params := captured["params"].(map[string]any)
assert.Equal(t, "tools/call", captured["method"])
assert.Equal(t, "session_log", params["name"])
args := params["arguments"].(map[string]any)
assert.Equal(t, "_routing", args["skill"])
// Bug B fix: the record carries the real skill so /pass-rate?skill=review sees it.
assert.Equal(t, "review", args["skill"])
assert.Equal(t, "decide", args["phase"])
assert.Equal(t, "skip", args["final_status"])
// Bug A fix: a successful routed call logs "pass", not "skip".
assert.Equal(t, "pass", args["final_status"])
assert.Contains(t, args["message"].(string), "review: local")
// session grouping is preserved via session_id.
assert.Equal(t, "sess-1", args["session_id"])
assert.Equal(t, "/home/x/proj", args["project_root"])
assert.Equal(t, float64(1234), args["duration_ms"])
@@ -59,23 +67,40 @@ func TestLoggerLogFailure(t *testing.T) {
}))
defer srv.Close()
l := routing.NewLogger(srv.URL)
l := routing.NewLogger(srv.URL, "test-token")
err := l.LogDecision(context.Background(), routing.LogEntry{
SessionID: "s", Skill: "debug", Decision: "local", Message: "litellm down", Failed: true,
})
require.NoError(t, err)
args := captured["params"].(map[string]any)["arguments"].(map[string]any)
assert.Equal(t, "debug", args["skill"])
assert.Equal(t, "fail", args["final_status"])
}
func TestLoggerOmitsAuthWhenTokenEmpty(t *testing.T) {
var authHeader string
hasAuth := false
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
authHeader = r.Header.Get("Authorization")
_, hasAuth = r.Header["Authorization"]
_ = json.NewEncoder(w).Encode(map[string]any{"jsonrpc": "2.0", "id": 1, "result": map[string]any{}})
}))
defer srv.Close()
l := routing.NewLogger(srv.URL, "")
require.NoError(t, l.LogDecision(context.Background(), routing.LogEntry{Skill: "review", SessionID: "_routing", Decision: "local"}))
assert.False(t, hasAuth, "no Authorization header should be set when token is empty")
assert.Equal(t, "", authHeader)
}
func TestLoggerSurfacesUpstreamError(t *testing.T) {
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
http.Error(w, "down", http.StatusBadGateway)
}))
defer srv.Close()
l := routing.NewLogger(srv.URL)
l := routing.NewLogger(srv.URL, "test-token")
err := l.LogDecision(context.Background(), routing.LogEntry{Skill: "x", SessionID: "y", Decision: "local"})
require.Error(t, err)
}
+2 -2
View File
@@ -50,7 +50,7 @@ func newRouter(t *testing.T, llm *fakeLLM, passRate float64) (*routing.Router, *
r := &routing.Router{
Fetcher: routing.NewFetcher(brain.URL, "7d", time.Minute),
Logger: routing.NewLogger(brain.URL),
Logger: routing.NewLogger(brain.URL, ""),
Policy: routing.Policy{Floor: 0.9, Ceil: 0.7},
FastModel: "koala/qwen35-9b-fast",
ThinkingModel: "iguana/gemma4-26b",
@@ -117,7 +117,7 @@ func TestRouterDefaultsToFastWhenBrainUnreachable(t *testing.T) {
llm := &fakeLLM{resp: "ok"}
r := &routing.Router{
Fetcher: routing.NewFetcher(brain.URL, "7d", time.Minute),
Logger: routing.NewLogger(brain.URL),
Logger: routing.NewLogger(brain.URL, ""),
Policy: routing.Policy{Floor: 0.9, Ceil: 0.7},
FastModel: "koala/qwen35-9b-fast",
ThinkingModel: "iguana/gemma4-26b",
+29 -59
View File
@@ -7,14 +7,14 @@ description: Disciplined end-of-session closeout for a Claude.ai chat before arc
Capture a finishing Claude.ai work session into durable storage before the chat is archived and its context is lost. The goal is simple and load-bearing: **after this runs, a fresh session (or another agent) can reconstruct what was decided, what was shipped, and what is still open — without the original chat.**
This skill is **batch**: one session in, findings out, done. It does not loop or re-read its own fresh output semantically (see Phase 5). Run the phases in order. Stop at any confirmation gate that says STOP.
This skill is **batch**: one session in, findings out, done. Run the phases in order. Stop at any confirmation gate that says STOP.
## Operating constraints (read first)
- **Gitea owner is always `mathias`.** Never guess another owner.
- **Ground-truth at HEAD before acting.** Issue bodies and doc references rot — stale hostnames, retired services, moved endpoints. Before closing/commenting on any issue, `gitea:issue_get` it fresh. Before asserting an infra fact, verify it; do not copy it from memory or from a stale issue body.
- **Current infra truths** (verify rather than trust, but these are the known-good baseline): Gitea is `git.d-ma.be` (not `gitea.d-ma.be`). LiteLLM is `http://koala:30401/v1/` (public `https://llm-api.d-ma.be`); piguard runs NGINX Proxy Manager only — never reference `piguard:4000` or `koala:4000`. Identity provider is Authentik (Dex migration complete).
- **Side-effects need a confirmation gate.** Closing issues, committing files, and writing to the brain are all real writes. Surface exactly what will happen and get a clear yes before doing it. Reads are free; writes are gated.
- **Side-effects need a confirmation gate.** Closing issues and capturing to brain/Gitea/ai-sessions are real writes. Surface exactly what will happen and get a clear yes before doing it. Reads are free; writes are gated.
- **Never fabricate.** If the session didn't produce a decision worth persisting, say so and skip that write. An empty-but-honest closeout beats an invented one.
## Phase 1 — Harvest
@@ -34,76 +34,46 @@ For every repo touched this session, get its true current state before proposing
Do not write anything in this phase. This is the read pass.
## Phase 3 — Confirm and act on issue changes
## Phase 3 — Plan the issue changes
Present a single consolidated plan of issue actions: which to close (with closing comment), which to file (discovered-but-deferred work — token-budget gaps, recorded limitations, v2 follow-ups), which to comment on. Include the exact title/body for any new issue and the closing rationale for any close.
Decide the issue actions: which to close (with closing comment), which to file (discovered-but-deferred work — token-budget gaps, recorded limitations, v2 follow-ups), which to comment on. Include the exact title/body for any new issue and the closing rationale for any close.
**GATE — STOP and get explicit confirmation before any issue write.** Issue closes and new issues are side-effects. Once confirmed, execute them (`gitea:issue_close`, `gitea:issue_create`, `gitea:issue_comment`, all owner `mathias`), correcting any rotted references you found in Phase 2 as you go.
These actions are **carried into the Phase 4 capture call** as `tickets[]` rather than executed here with direct `gitea:issue_*` calls — routing them through capture puts each one into the I5 audit record. (Closing an issue that needs a separate explanatory comment first is the one case to do directly; otherwise prefer the capture path.)
## Phase 4 — Commit the canonical session summary
## Phase 4 — Capture (one uniform call)
Write one summary file to `mathias/ai-sessions`, committed directly to `main` via `gitea:file_write_branch` (no PR — this repo is solo and unprotected; if branch protection is ever added, fall back to a branch + PR).
Persist the session via a **single `capture` call** (the `brain:capture` MCP tool, live on the Claude.ai connector). Capture owns the writes server-side — insights → brain, action items → Gitea tickets, summary → ai-sessions — plus the I1 sovereignty gate, the I5 audit record, and the supersession/read-after-write discipline. The skill's job is to *assemble the payload*, not to write each store itself. Do NOT fall back to separate `gitea:file_write_branch` + `brain_write` steps unless `capture` is unreachable (see fallback below).
**Path:** `summaries/claudeai/<YYYY-MM>/<YYYY-MM-DD>-<topic-slug>-<chatid8>.md`
where `<chatid8>` is the first 8 chars of the chat's UUID if known, else a short stable slug. `claudeai` has no host segment — Claude.ai is Anthropic-side, not a homelab host.
**Assemble one payload:**
**Frontmatter — the REDUCED live-capture schema.** A live close-session capture cannot populate the batch-export telemetry (token counts, message counts, duration_ms, permission_mode) — those only exist in the account export pipeline. Write only what's truthfully known, and mark fidelity so a reader (or the batch pipeline) can tell a live capture from an export:
- **`insights[]`** — the generalizable learnings from Phase 1 (decisions/failures worth re-reading). Each: `{text, wing, hall}`; add `supersede_slug` to revise a prior note in place instead of creating a duplicate. `hall` ∈ facts/decisions/failures/hypotheses/sources.
- **`tickets[]`** — the issue actions from Phase 3: `{repo, action, ...}` where action ∈ create/close/comment. Owner is always `mathias` (server-forced).
- **`summary`** — `{title, body, repos_touched}`. Capture writes it to `ai-sessions` and stamps `fidelity` in frontmatter. Body stays reconstructable: one-paragraph summary, decisions, key artifacts, open threads.
- **`context`** — `{harness: "claudeai-chat", session_ref: <chatid8-or-slug>, fidelity: "live-capture", actor: "mathias", classification: <see gate below>}`.
```yaml
---
title: "<concise session title>"
client: "claudeai"
interface: "claudeai-chat"
date: "<YYYY-MM-DD>"
repos_touched: [<repo slugs>]
topic_tags: [<tags>]
outcome: "<shipped|in-progress|abandoned>"
fidelity: "live-capture" # NOT an export; reconstructed live from chat
captured_by: "close-session-skill"
---
```
**THE CLASSIFICATION GATE (read before calling — this is where capture refuses).**
Capture computes an **effective classification = the strictest across EVERY target it touches** (each insight's `wing`, each ticket's `repo`, and every entry in `summary.repos_touched`), then refuses if that effective level is `confidential` and the origin is us-nexus (claude.ai is us-nexus). Levels come from `classification.yaml` at the brain root (source of truth, #67), with the code defaults as the floor: `hyperguild`/`homelab` → internal; `client-*` → confidential; **anything untagged → confidential (fail-safe)**.
- **Tagged `internal` today** (safe through claude.ai): wings `hyperguild`, `homelab`; repos `brain`, `ai-sessions`, `infra`, `hyperguild`, `homelab`, `tapir`, `agentsquad`, `jepa-fx-risk`, `swedsl`. Treat `classification.yaml` as authoritative — this list is a hint, not gospel.
- Declare `context.classification: "internal"` for normal homelab work.
- `summary.repos_touched`, insight `wing`s, and ticket `repo`s are classification INPUTS, not free-form metadata — every target must resolve `internal` or the whole capture escalates to `confidential` and the gate refuses via claude.ai. Listing the central homelab repos (incl. `brain`/`ai-sessions`) is now fine; they're tagged. The summary always lands in `ai-sessions` (internal), so the summary path itself never escalates.
- If a session genuinely touched **`client-*` or otherwise-untagged** material, it cannot be captured through claude.ai — note that in the verdict rather than trying to force it.
Do not invent the export-only fields. `fidelity: live-capture` is the honest signal; if the batch export later produces a richer summary for the same session, the export is source of truth and supersedes this.
**GATE — dry-run first, then execute.**
1. Call `capture` with `dry_run: true`. It validates the whole payload and returns the would-be receipt + `effective_classification`, writing nothing.
2. **STOP. Show the dry-run receipt** (effective classification, the insights/tickets/summary that would land) and get explicit confirmation.
3. On confirmation, call `capture` again with `dry_run: false`. Read the returned receipt: it is partial-aware (`errors[]`, per-item `ok`). Report exactly what landed.
**Body** (keep it reconstructable, not exhaustive):
```markdown
## One-paragraph summary
## Decisions
## Key artifacts
## Open threads
```
If `capture` is **unreachable** (tool not on the connector — e.g. a session that started before a deploy; a tool-list refresh usually fixes it): say so. Only then fall back to the legacy inline path (`gitea:file_write_branch` summary + `brain_write`/`brain_update` + `brain_get` confirm), and note in the verdict that the I5 audit record was NOT produced.
**GATE — STOP, show the full file (path + frontmatter + body), get explicit confirmation before committing.**
## Phase 5 — Brain orientation note (the durable "where we are" record)
Write one brain note so a fresh session can orient without the chat. This uses the `brain_update`/`brain_get` verbs (live since 2026-06).
**Target:** `wing: <domain>` (the project/topic domain, e.g. `hyperguild`, `jepa-fx`), `hall: decisions`. The note is a knowledge-type record (a decision/orientation), grouped by knowledge-type, not by interface surface.
**Batch read-after-write discipline (important — do these in order, do not interleave):**
1. **Read first, before any write.** Check whether an orientation note already exists for this wing/topic. Do your "does this already exist / what should I supersede" reads NOW, up front. BM25/keyword search and `brain_get` are immediate; semantic/vector search may lag up to ~5 min after a write, so never rely on a semantic query to find something you wrote earlier in this same run.
2. **Write or supersede:**
- **New note** → `brain_write` (wing, hall: decisions). Returns `{id, path, content_hash}`.
- **Superseding a prior orientation note** → `brain_update` (slug or path, wing, hall, content, reason). Whole-note replace; stamps `supersedes`/`updated_at`; returns `{id, path, content_hash, superseded}`. Use this instead of a second `brain_write` to the same slug — blind re-write creates duplicates/contradictions, which is the exact failure brain_update exists to prevent.
3. **Confirm it landed** via `brain_get(id)` and check the returned `content_hash` matches what the write returned. This is the read-after-write confirmation — do it with `brain_get`, never a semantic query.
**RULE: no semantic/vector brain query after the first `brain_update` in this run.** The batch shape makes this natural — read up front, write, confirm by id. If you ever find the skill wanting to semantic-search a just-superseded note, stop and flag it (that's the signal the staleness window matters and needs the synchronous-reembed follow-up).
**GATE — STOP, show the note (target wing/hall, new-vs-supersede, full content), get explicit confirmation before the brain write.**
After the note lands, if it relates to a note in another wing, create the cross-link inline with `brain_tunnel(source, target)` (idempotent; both paths brain-relative, must be in different wings). Optionally append a `session_log` entry (`session_id`, `skill: close-session`, `phase`, `final_status`) for telemetry. Both are now callable directly from Claude.ai — no Claude Code/Crush handoff needed.
## Phase 6 — Verdict
## Phase 5 — Verdict
Deliver a final "safe to archive" verdict in the chat. Either:
- **SAFE TO ARCHIVE** — list what landed (issues closed/filed with numbers, summary path, brain note id, any tunnels) so the trail is auditable. Then list anything still in the user's queue (e.g. a PR awaiting their merge, a decision owed next session).
- **NOT YET** — name the specific gate that wasn't passed or the write that failed, and what to do about it.
- **SAFE TO ARCHIVE** — list what landed from the capture receipt (issues closed/filed with numbers, summary path, brain note ids/paths) so the trail is auditable. Then list anything still in the user's queue (e.g. a PR awaiting their merge, a decision owed next session).
- **NOT YET** — name the specific gate that wasn't passed, the capture refusal reason, or the per-item error from the receipt, and what to do about it.
Never claim safe-to-archive if any gated write was declined or errored. The verdict is the skill's contract: if it says safe, the session can be lost without losing the work.
Never claim safe-to-archive if the capture refused, any receipt item errored, or a gated confirmation was declined. The verdict is the skill's contract: if it says safe, the session can be lost without losing the work.
## Why the gates and the batch discipline matter
## Why the gate and the single-call shape matter
The whole point is durability across a context reset. Every gate is a place where a wrong write would silently corrupt the record (close the wrong issue, overwrite a good brain note, commit a half-truth). The batch read-discipline in Phase 5 exists because the brain's vector index refreshes out-of-band: write-then-semantically-reread in the same run can read stale, so the skill front-loads reads and confirms writes by id. Get those right and the skill does what it promises — nothing important is lost when the chat goes away.
The whole point is durability across a context reset. The capture call is the one place a wrong payload would silently corrupt the record (close the wrong issue, escalate to a refusal, commit a half-truth), which is why it is dry-run-then-confirm. Routing everything through one `capture` keeps the supersession discipline, the read-after-write confirmation, and the I5 audit trail server-side — the skill never has to carry those rules itself, and every closeout is uniformly audited. Get the payload and the classification right and the skill does what it promises — nothing important is lost when the chat goes away.