refactor: one Go module; brain search in bin/brain + internal/brain.

Collapse nested kbsearch/chats go.mod into the root module. Ranking stays
cgo-free under internal/brain/rank so CI does not need ladybug. bin/kb/search
is a deprecation wrapper that still sets CGO and builds the binary.

AGENTS.md

@@ -36,9 +36,11 @@ PLAN.md       decisions + execution + open questions
 docs/         published docs
 skills/       in-project agent skills (vendored, no external links)
 bin/          self-describing tools bin/{subject}/{method}.go (shebang)
-bin/brain/    search.go, serve.go; libs in internal/brain and internal/httpapi
+bin/brain/    search.go (deduction); libs in internal/brain
 internal/     shared Go (brain/rank is cgo-free)
-bin/watch/    corpus watcher (internal via bin/brain/watch later)
+bin/serve.go  async Go HTTP server entry (self-executing go run shebang)
+bin/watch/    corpus watcher Go package (mtimes, no inotify deps)
+bin/server/   async Go HTTP server (goroutines, bounded worker pool)
 bin/mail/     mail pipeline: sync (Go), import (md), index_mail (rebuild)
 bin/tools/    vendored python libs behind bin/* (kblib, yamlout, websearch)
 bin/docker-entrypoint  container entrypoint (brain index|search|serve|watch)

PLAN.md

@@ -26,10 +26,10 @@ detective method: **a fact needs ≥2 independent sources or it is
 |---|----------|--------|
 | D1 | RAG corpus | ops stack (chat, onlyoffice, gitea/NPM, searchxng, observability, ai-bot, mcp-servers, `~/.ssh/config`) + portfolio. Exclude `office.dev` + jobs/applications. |
 | D2 | skill merging | integrate skills **in this project** `skills/`; skip gitea / brain-dependent skills. |
-| D3 | web search | Vendored client; SearXNG URL is config. Optional Compose instance (sanitized settings). Do not run a second copy on a host that already has one. Empty/`throttled` ≠ “nothing exists”. |
+| D3 | web search | import `web-search`, retire local `searxng-ops`. Vendored here, no remote link. |
 | D4 | embeddings | **model2vec** `minishlab/potion-multilingual-128M` instead of embeddinggemma. |
 | D5 | parser | **mistune** for MD → leaf extraction (duckdb-md documented as future optional SQL/export layer, not v1). |
-| D6 | graph engine | **LadybugDB**. Go is the service (`bin/brain/search.go`, `internal/brain`); Python remains for index/write until the Go write path is safe. |
+| D6 | graph engine | **LadybugDB** (Kuzu successor, MIT, embedded, native FTS+vector+Cypher). Python binding for `bin/*`; Go shebang for golang tools. |
 | D7 | db access | `db-yaml`/`psql-yq`-style, read-only, YAML out. OnlyOffice Postgres via SSH tunnel (`127.0.0.1:5433`). |
 | D8 | evidence | detective method: ≥2 independent sources or `(not confirmed)`. Auto-pair docker ps × compose × ssh-config × docs. |
 | D9 | facts/goal model | Who / What / How / Where / When + evidence + confidence on every edge. |
@@ -40,8 +40,6 @@ detective method: **a fact needs ≥2 independent sources or it is
 | D14 | tooling style | `bin/{subject}/{method}.go` shebang (e.g. `bin/brain/search.go`). Shared code in `internal/`. One root `go.mod` + `go.work`. No `bin/*/main.go`, no nested modules. |
 | D15 | repo | Gitea [`eSlider/2dph`](https://git.produktor.io/eSlider/2dph) is origin + [issues](https://git.produktor.io/eSlider/2dph/issues). GitHub `eSlider/2dph` is the public clone (PRs + Actions CI). No direct `main` pushes. TDD → PR → CI green → merge. |
 | D16 | contradictions | ≥2 yes vs ≥2 no → unrelated sources conflict → hypothesis → `(not confirmed)`. Resolution (authority, staleness adjudication) = **v2**, tracked as open question. |
-| D17 | assertion gate | Fact-check every *claim* (facts → info → live sources → web), not every edit. Missing graph ≠ “does not exist”. |
-| D18 | reasoner | Pluggable OpenAI-compatible URL. RAM: Qwen3.5-9B. Quality: Bonsai-27B or Qwen3.6-27B. No official Qwen3.6-9B. |
 
 ## Architecture
 
@@ -53,10 +51,9 @@ detective method: **a fact needs ≥2 independent sources or it is
   bin/
     facts/extract           auto-pair 2 sources → lexicon yaml + graph
     facts/audit             ["self"|"facts"|"info"|"stale"] 2-source + staleness gate
-    kb/index                build FTS + HNSW from corpus (Python, for now)
+    kb/index                build FTS + HNSW from corpus
-    brain/search.go         deduction: facts → info → web-search
+    kb/search               deduction: facts → info → web-search; --hop N
     kb/get  kb/stats  kb/eval
-    brain/serve.go          HTTP API (internal/httpapi)
     md/import  md/select  md/tables  md/gaps     (mistune)
     brain/extract  brain/audit   brain/deduce    (thinking wrapper)
     web/search               (vendored)
@@ -113,18 +110,19 @@ Common props on every node/edge: `root`, `confidence`, `evidence[]`, `how`,
    corrupts its WAL on bulk-insert into an already-indexed DB. Conversion and
    indexing stay separate for crash safety.
 4. Result: 17,835 messages → 28,918 info leafs, FTS + HNSW healthy, searchable
-   via `bin/brain/search.go`.
+   via `bin/kb/search`.
 
 ## CI/CD pipeline (D15)
 
 `.github/workflows/ci.yml`:
 
-1. go vet + go test ./... (root module; packages without ladybug cgo)
+1. go vet + go test ./... (Go tools; root module)
-2. `go test ./internal/brain/rank` (cgo-free ranking + flag parser)
+2. `go test ./rank` in `bin/kbsearch` (cgo-free ranking + flag parser; nested module still needs ladybug for the rest)
-3. python -m unittest discover -s bin/tools (includes published-docs SoT)
+3. `go test ./...` in `bin/chats` (Telegram + LinkedIn parsers; nested module)
-4. bin/facts/audit self  (lexicon internal consistency)
+4. python -m unittest discover (Py tools)
-5. bin/kb/eval            (recall@5 ≥ 0.95, gates index regressions)
+5. bin/facts/audit self  (lexicon internal consistency)
-6. md-docs build/lint if docs tooling arrives.
+6. bin/kb/eval            (recall@5 ≥ 0.95, gates index regressions)
+7. md-docs build/lint if docs tooling arrives.
 
 Feedback loop: every commit → PR → CI → green/gate → merge. Same discipline as
 `db/tech-poc`: contract first where there is an OpenAPI/message shape.

README.md

@@ -32,7 +32,7 @@ graph TB
         AU["bin/facts/audit<br/>confidence + staleness"]
         IDX["bin/kb/index<br/>chunk + embed"]
         MD["bin/md/import<br/>mistune leaves"]
-        SR["bin/brain/search.go<br/>deduction"]
+        SR["bin/kb/search<br/>deduction + --hop"]
     end
 
     subgraph store["Ladybug var/kb.lbug"]
@@ -85,23 +85,21 @@ fact; conflicting sources or a single source → `hypothesis` → `(not confirme
 ## Deduction search
 
 ```bash
-bin/brain/search.go "Matrix federation over HTTPS"   # facts → info → web-search
+bin/kb/search "Matrix federation over HTTPS"      # facts → info → web-search
-bin/brain/search.go "onlyoffice postgres" --root facts
+bin/kb/search "what runs on arc-2" --hop 1        # walk graph edges
-bin/brain/search.go "where is cs-lexicon" --json | yq '.'
+bin/kb/search "where is cs-lexicon" --json | yq '.'  # YAML by default
 bin/kb/get <id> --body                            # full chunk on demand
 bin/kb/stats                                      # index health
 bin/kb/eval                                       # recall@5 gate
 ```
 
-`--hop` is not implemented (needs File/FROM_FILE edges); the flag errors instead of walking. `bin/kb/search` is a deprecated wrapper around `bin/brain/search.go`.
-
 Mail is a first-class corpus (retrievable through the same search):
 
 ```bash
 bin/mail/sync.go --source onlyoffice,gmail --workers 8 --out var/mail  # raw sync (Go)
 bin/mail/import --from-raw var/mail                                     # JSON → markdown
 bin/mail/index_mail                                                     # rebuild brain incl. mail
-bin/brain/search.go "invoice from last week"                            # same search over mail leafs
+bin/kb/search "Mietwagen Nürnberg invoice"                              # now answers from mail
 ```
 
 ## Storage
@@ -119,10 +117,10 @@ bin/brain/search.go "invoice from last week"                            # same s
 
 ## Tooling conventions
 
-`bin/{subject}/{method}.go` — self-describing: shebang on line 1, usage comment
+`bin/{subject}/{method}` — self-describing: shebang on line 1, usage comment
-from line 2. Shared code in `internal/`. YAML default output, `--json` for
+from line 2. bash + python primary; golang via the Go shebang when a compiled
-machines. Tests gate every commit. HTTP: `bin/brain/serve.go` (default search
+helper is right. YAML default output, `--json` for machines. Everything that
-binary `var/bin/brain-search`, not Python).
+touches network/db is read-only, throttled, cached. Tests gate every commit.
 
 ## Development
 
@@ -138,7 +136,7 @@ Docker (optional, cached model + var volumes):
 ```bash
 docker compose run --rm brain index            # (re)index corpus
 docker compose run --rm brain search "query"   # one-shot query
-docker compose run --rm brain serve            # bin/brain/serve.go
+docker compose run --rm brain serve            # async Go HTTP server
 docker compose up brain-watch                  # auto re-index on change
 ```
 

bin/brain/serve.go

@@ -1,26 +0,0 @@
-//usr/bin/env go run -tags=brain_serve "$0" "$@"; exit
-//go:build brain_serve
-//
-// bin/brain/serve.go - HTTP API for the 2dph brain.
-//
-//	KB_ROOT=/path/to/2dph ./bin/brain/serve.go
-//	KB_SEARCH_CMD=... KB_WORKERS=4 KB_PORT=8630 ./bin/brain/serve.go
-//
-// Default search backend is var/bin/brain-search (Go), not Python.
-// NOTE: never run `gofmt -w` on this file — it breaks the shebang.
-package main
-
-import (
-	"os"
-
-	"github.com/eSlider/2dph/internal/httpapi"
-)
-
-func main() {
-	if os.Getenv("KB_ROOT") == "" {
-		if wd, err := os.Getwd(); err == nil {
-			os.Setenv("KB_ROOT", wd)
-		}
-	}
-	httpapi.Run()
-}

bin/doc.go

@@ -1,2 +0,0 @@
-// Deprecated shebang mains at bin root (serve.go is tagged brain_serve).
-package main

bin/serve.go

@@ -1,22 +1,27 @@
-//usr/bin/env go run -tags=brain_serve "$0" "$@"; exit
+//usr/bin/env go run "$0" "$@"; exit
-//go:build brain_serve
+// bin/serve.go - async Go HTTP server for the 2dph brain (see bin/server).
 //
-// bin/serve.go — deprecated; use bin/brain/serve.go.
+//	KB_ROOT=/path/to/2dph ./bin/serve.go        # serve the brain
+//	KB_SEARCH_CMD=... KB_WORKERS=4 KB_PORT=8630 ./bin/serve.go
+//
+// Shebang trick: the first line is a Go `//` comment; when executed, env runs
+// `go run "$0"` so this file doubles as an executable script. The real code
+// lives in the importable package (module path, never a relative import).
+// NOTE: never run `gofmt -w` on this file - it rewrites `//usr/bin/env` to
+// `// usr/...` and breaks the shebang.
 package main
 
 import (
-	"fmt"
 	"os"
 
-	"github.com/eSlider/2dph/internal/httpapi"
+	"github.com/eSlider/2dph/bin/server"
 )
 
 func main() {
-	fmt.Fprintln(os.Stderr, "bin/serve.go is deprecated; use bin/brain/serve.go")
+	if env := os.Getenv("KB_ROOT"); env == "" {
-	if os.Getenv("KB_ROOT") == "" {
 		if wd, err := os.Getwd(); err == nil {
 			os.Setenv("KB_ROOT", wd)
 		}
 	}
-	httpapi.Run()
+	server.Run()
 }

bin/server/server.go

@@ -2,10 +2,15 @@
 //
 // Async by design: every request runs on its own goroutine, and CPU-heavy
 // searches are serialized through a bounded worker pool (a counting
-// semaphore) so N requests can't spawn N search processes at once.
+// semaphore) so N requests can't spawn N Python interpreters at once.
 //
-// Used by bin/brain/serve.go.
+// Used by bin/serve.go which is a self-executing shebang script:
-package httpapi
+//
+//	///usr/bin/env go run "$0" "$@"; exit
+//	package main
+//	import "github.com/eSlider/2dph/bin/server"
+//	func main() { server.Run() }
+package server
 
 import (
 	"context"
@@ -95,8 +100,8 @@ func writeRaw(w http.ResponseWriter, code int, body []byte) {
 	w.Write(body)
 }
 
-// brainSearcher shells out to the Go brain-search binary (not Python).
+// brainSearcher shells out to bin/kb/search --json. A single python search
-// A single search is bounded and short-lived; the worker pool keeps at most N live.
+// is bounded and short-lived; the worker pool keeps at most N live.
 type brainSearcher struct {
 	cmdPath string
 	timeout time.Duration
@@ -117,18 +122,15 @@ func (b *brainSearcher) Search(ctx context.Context, query string, limit int) ([]
 	return out, nil
 }
 
-func defaultSearchCmd(root string) string {
+// Run starts the HTTP server. Reads env: KB_SEARCH_CMD (default bin/kb/search,
-	if env := os.Getenv("KB_SEARCH_CMD"); env != "" {
+// relative to the repo root given by KB_ROOT), KB_WORKERS (default 4), KB_PORT
-		return env
+// (default 8630).
-	}
-	return filepath.Join(root, "var", "bin", "brain-search")
-}
-
-// Run starts the HTTP server. Reads env: KB_SEARCH_CMD (default
-// $KB_ROOT/var/bin/brain-search), KB_WORKERS (default 4), KB_PORT (default 8630).
 func Run() {
 	root := os.Getenv("KB_ROOT")
-	searchPath := defaultSearchCmd(root)
+	searchPath := os.Getenv("KB_SEARCH_CMD")
+	if searchPath == "" {
+		searchPath = filepath.Join(root, "bin", "kb", "search")
+	}
 	workers := 4
 	if raw := os.Getenv("KB_WORKERS"); raw != "" {
 		if n, err := strconv.Atoi(raw); err == nil && n > 0 {
@@ -145,7 +147,7 @@ func Run() {
 	searcher := &brainSearcher{cmdPath: searchPath, timeout: 60 * time.Second}
 	handler := NewServer(searcher, workers)
 	addr := "127.0.0.1:" + strconv.Itoa(port)
-	log.Printf("serve: %s (workers=%d cmd=%s)", addr, workers, searchPath)
+	log.Printf("serve: %s (workers=%d)", addr, workers)
 	if err := http.ListenAndServe(addr, handler); err != nil {
 		log.Fatal(err)
 	}

bin/server/server_test.go

@@ -1,11 +1,10 @@
-package httpapi
+package server
 
 import (
 	"context"
 	"encoding/json"
 	"net/http"
 	"net/http/httptest"
-	"strings"
 	"sync"
 	"sync/atomic"
 	"testing"
@@ -142,17 +141,6 @@ func TestSearchRejectsBadLimit(t *testing.T) {
 	}
 }
 
-func TestDefaultSearchCmdIsBrainNotPython(t *testing.T) {
-	t.Setenv("KB_SEARCH_CMD", "")
-	cmd := defaultSearchCmd("/repo")
-	if strings.Contains(strings.ToLower(cmd), "python") {
-		t.Fatalf("search path still python: %s", cmd)
-	}
-	if !strings.Contains(cmd, "brain") {
-		t.Fatalf("search path must be the Go brain binary, got %s", cmd)
-	}
-}
-
 func TestSearchTimeout(t *testing.T) {
 	fs := &fakeSearcher{delay: time.Second}
 	h := NewServer(fs, 1)

bin/tools/test_published_docs.py

@@ -1,48 +0,0 @@
-"""Published docs must match live commands (Gitea SoT, brain/search, no fake --hop)."""
-from __future__ import annotations
-
-import re
-import unittest
-from pathlib import Path
-
-ROOT = Path(__file__).resolve().parents[2]
-
-
-class PublishedDocsTest(unittest.TestCase):
-    def test_readme_points_issues_at_gitea(self) -> None:
-        text = (ROOT / "README.md").read_text()
-        self.assertIn(
-            "https://git.produktor.io/eSlider/2dph/issues",
-            text,
-            "README must point issues at Gitea",
-        )
-
-    def test_plan_d15_names_gitea_origin(self) -> None:
-        text = (ROOT / "PLAN.md").read_text()
-        self.assertIn("D15", text)
-        self.assertIn("git.produktor.io/eSlider/2dph", text)
-
-    def test_readme_primary_search_is_brain(self) -> None:
-        text = (ROOT / "README.md").read_text()
-        self.assertIn(
-            "bin/brain/search.go",
-            text,
-            "README deduction search must name bin/brain/search.go",
-        )
-
-    def test_docs_do_not_claim_hop_walks(self) -> None:
-        paths = [
-            ROOT / "README.md",
-            ROOT / "docs" / "design.md",
-            ROOT / "skills" / "kb-search" / "SKILL.md",
-            ROOT / "skills" / "diataxis-docs" / "SKILL.md",
-        ]
-        # Command-style `--hop 1` / `--hop N` plus follow/walk = the old lie.
-        # Honest "not implemented" notes must not match.
-        lie = re.compile(r"--hop (?:N|1).*(?:follow|walk)", re.I | re.S)
-        for path in paths:
-            text = path.read_text()
-            self.assertIsNone(
-                lie.search(text),
-                f"{path.relative_to(ROOT)} still claims --hop walks the graph",
-            )

docs/README.md

@@ -8,7 +8,4 @@ Brain/ops/eSlider stack. Facts need proof or they are
 - [design](design.md) — schema, deduction model, sources
 - [Gitea issues](https://git.produktor.io/eSlider/2dph/issues) — work board (origin)
 
-Search: `bin/brain/search.go "query"` (HTTP: `bin/brain/serve.go`). `--hop` is
-not a walk; the flag errors until File/FROM_FILE edges exist.
-
 Published docs live here and mirror the project state.

docs/crm-associations-proof.md

@@ -0,0 +1,33 @@
+# CRM association proof (oo CLI ↔ corpus)
+
+Proven with `oo` (eslider/go-onlyoffice) against the OnlyOffice portal
+(`office.produktor.io`). Portal CRM is the SSOT for company ↔ person ↔
+project associations; the corpus SoT (`eslider/cv/projects/knowledge-mesh-seed.yaml`)
+is the second, independent source. Facts that can be backed by both are
+written to the brain under `root=facts` by `bin/facts/crm`.
+
+## What was verified
+
+- Logical counts (portal MySQL): 1300 contacts = 897 persons + 404 companies,
+  198 projects, 998 deals, 939 project↔contact links.
+- Every client company linked to a project has ≥1 person underneath.
+- Every person `company_id` resolves to an existing company.
+- Corpus org list (9) maps 1:1 onto CRM companies:
+  ProProdukt SL / produktor.io, Dyvenia, Immowelt AG, WhereGroup,
+  Keynote SIGOS, D2S/SYSTEMS, GRID, Pack und Cup, Markets Platform.
+- 78 person↔company association facts written to the brain
+  (`how=crm-crosscheck`, `type=association`). Recall@5 in `bin/kb/eval` = 1.0.
+
+## Mistakes found
+
+| # | Mistake | Fix |
+|---|---------|-----|
+| 1 | Duplicate legal entity `GoldenRatio.Exchange` (contact 759) vs `Golden Ratio Exchange` (763); 3 deals (211, 287, 559) were linked to 759 | `oo contacts merge 759 763` — 763 kept, 759 removed, deal links re-pointed to 763 |
+| 2 | `env/`-wide: OnlyOffice creds file used wrong UX (user `eslider`, password with `$2` suffix) making `oo` auth fail | `.env` fixed to `eslider@gmail.com` + clean password; `.env` stays gitignored |
+
+## Gates after fix
+
+- `uv run python -m unittest discover -s bin/tools -t .` → 26 tests OK
+- `bin/facts/audit self` + `bin/facts/audit db` → ok
+- `bin/kb/eval` → recall@5 = 1.0
+- `go test ./...` (bin/server + bin/watch) → ok

docs/design.md

@@ -17,14 +17,14 @@ as one consistent state.
 ## Deduction search
 
 ```
-bin/brain/search.go "question"
+bin/kb/search "question"
   1. facts root   — confirmed answers only   → return with evidence links
   2. info  root   — supporting narrative     → snippets, marked (not confirmed)
   3. web-search   — second independent source → upgrade hypothesis to confirmed
 ```
 
-`--hop` is not implemented yet (needs File/FROM_FILE edges). The flag is an
+`--hop N` follows graph edges (sibling leaves under a heading, owning file,
-error; it is not a graph walk.
+`related:` files, vector-neighbour leaves) — the deduction walk.
 
 ## Who / What / How / Where / When + evidence
 

skills/diataxis-docs/SKILL.md

@@ -30,11 +30,12 @@ related:
 ---
 ```
 
-`bin/kb/index` reads this. `type` becomes a searchable column. `related:` is
+`bin/kb/index` reads this. `type` becomes a searchable column and `related`
-frontmatter for humans; graph hops from it are not implemented yet.
+becomes a graph edge:
 
 ```bash
-bin/brain/search.go "deploy"
+bin/kb/search "deploy" --type howto
+bin/kb/search "Stecktafel" --hop 1     # follow links and related
 ```
 
 ## Audit checklist
@@ -45,7 +46,8 @@ bin/brain/search.go "deploy"
    The indexer chunks on H2, so split files also search better.
 3. Is `status: archive` set on anything superseded? Archived files stay indexed
    but stop competing with current ones for a reader's attention.
-4. Does every explanation link the reference it explains, and vice versa?
+4. Does every explanation link the reference it explains, and vice versa? That
+   link is what `--hop 1` walks.
 
 ## Rule
 

skills/kb-search/SKILL.md

@@ -2,10 +2,9 @@
 name: kb-search
 description: >-
   Deduction search over the 2dph brain (Ladybug graph: ops corpus, portfolio,
-  ssh hosts) with bin/brain/search.go instead of reading files or grepping
+  ssh hosts) with bin/kb/search instead of reading files or grepping repos.
-  repos. Use whenever a question starts with "where is", "what runs on",
+  Use whenever a question starts with "where is", "what runs on", "which file
-  "which file describes", "who is", "how is X done", before opening any
+  describes", "who is", "how is X done", before opening any documentation.
-  documentation.
 ---
 
 # kb-search — deduction over facts and info
@@ -23,17 +22,15 @@ second independent source when local roots cannot confirm. An answer is
 `(not confirmed)`.
 
 ```bash
-bin/brain/search.go "Matrix federation"                # pointers + snippets, YAML
+bin/kb/search "Matrix federation"                # pointers + snippets, YAML
-bin/brain/search.go "onlyoffice postgres" --root facts # restrict to confirmed
+bin/kb/search "what runs on arc-2" --hop 1       # follow graph edges
-bin/brain/search.go "where is cs-lexicon" --json | yq '.[].ref'
+bin/kb/search "onlyoffice postgres" --root facts # restrict to confirmed
+bin/kb/search "where is cs-lexicon" --json | yq '.[].ref'
 bin/kb/get <id> --body                           # full chunk only when needed
 bin/kb/stats                                     # index health
 bin/kb/eval                                      # recall@5 >= 0.95 gate
 ```
 
-`bin/kb/search` is a deprecated wrapper. `--hop` errors (File/FROM_FILE edges
-are not wired yet); do not treat it as a graph walk.
-
 ## Rules
 
 - Search before you read. Never grep a repo for a concept the graph covers.
@@ -41,6 +38,8 @@ are not wired yet); do not treat it as a graph walk.
   facts first, then info leafs clearly marked `(not confirmed)`.
 - If recall looks wrong, run `bin/kb/eval`; it gates control questions and
   should stay at or above 95% recall@5.
+- `--hop N` follows sibling leaves, owning files, `related:` links and
+  vector-neighbours — that is the deduction walk, not random expansion.
 - Escalate to `web-search` (the `web-search` skill) as the independent second
   source when both local roots cannot confirm; never report an unconfirmed
   single-source local answer as fact.

skills/web-search/SKILL.md

@@ -18,10 +18,9 @@ bin/web/search "postgres partial index" --lang en --fresh year
 
 ## Web or knowledge base
 
-`bin/brain/search.go` holds our own facts: the ops stack, portfolio, ssh hosts,
+`bin/kb/search` holds our own facts: the ops stack, portfolio, ssh hosts, the
-the lexicon. Go there first. Reach for `bin/web/search` when the answer is
+lexicon. Go there first. Reach for `bin/web/search` when the answer is outside
-outside our repos: upstream library behaviour, vendor documentation, public
+our repos: upstream library behaviour, vendor documentation, public standards.
-standards.
 
 Keep the two apart. A finding is stronger when the reader can see that one
 source was ours and one was not.