New command: gbrain providers [list|test|env|explain]. Explain emits
a schema_version:1 JSON matrix (agent-friendly). Auto-detects env
keys + probes localhost:11434 /v1/models (validates JSON shape, not
just port-open). Recommends the best provider with one-line reasoning.

gbrain init flags: --embedding-model provider:model (verbose) or
--model provider (shorthand, picks recipe default). Plus
--embedding-dimensions and --expansion-model. AI config flows into
saved GBrainConfig; engine.connect() configures gateway before
initSchema so vector column gets right dim.

config.ts: adds embedding_model, embedding_dimensions, expansion_model,
provider_base_urls. loadConfig() reads env vars but NEVER mutates
process.env — global-state leakage would break MCP, multi-brain, and
long-running workers.

cli.ts: routes 'providers' subcommand (CLI_ONLY, no engine needed);
connectEngine() calls configureGateway() before engine.connect().

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…(v0.15.0)

28 new unit tests across 4 files:

- test/ai/gateway.test.ts — 13 tests covering isAvailable() matrix
  for the silent-drop regression surface. Critical case: Gemini
  available when GOOGLE_GENERATIVE_AI_API_KEY set AND OPENAI_API_KEY
  absent. Pre-v0.15 brains silently dropped vectors in this config.
- test/ai/silent-drop-regression.test.ts — 3 source-level grep tests
  enforcing !process.env.OPENAI_API_KEY cannot re-enter the codebase
  at any of the three known sites.
- test/ai/schema-templating.test.ts — 4 tests for dim/model
  substitution in getPGLiteSchema() + PGLITE_SCHEMA_SQL back-compat.
- test/ai/config-no-env-mutation.test.ts — regression guard ensuring
  loadConfig() does not mutate process.env (Codex review C3).

All 28 pass locally. Existing unit suite (1397) + Tier 1 E2E (129)
+ Tier 2 skills E2E (3) all green against real Postgres+pgvector
and real OpenAI/Anthropic/openclaw.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Adds AI SDK deps (ai, @ai-sdk/openai, @ai-sdk/google,
@ai-sdk/anthropic, @ai-sdk/openai-compatible, zod, gray-matter,
eventsource-parser).

Note: Version jumped from 0.13.0 to 0.15.0 because upstream master
shipped 0.14.x (doctor DRY detection, Knowledge Runtime) while this
branch was in development. Keeping 0.15.0 as the natural next
release number for the AI providers cathedral.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.14.0 + v0.14.1 (doctor DRY detection + --fix auto-repair,
Knowledge Runtime, resolvers, integrity, shell jobs) into the embedding
providers branch.

Conflicts resolved:
- VERSION — kept v0.15.0 (this branch); upstream is v0.14.1
- package.json — v0.15.0 wins
- CHANGELOG.md — v0.15.0 entry placed above upstream's v0.14.1/v0.14.0/v0.13.1
- src/cli.ts — CLI_ONLY set merges both new commands; routing includes
  providers (this branch) + resolvers + integrity (upstream)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

CI failure: test hardcoded /Users/garrytan/... absolute paths that obviously
don't exist outside my machine. Resolve paths relative to import.meta.dir
so the test works on any checkout + in GitHub Actions.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.14.2 (#259 — 8 root-cause fixes from /investigate) into
the embedding-providers branch. Test count grows 1717 → 1766 on merge.

Conflicts resolved:
- VERSION — kept 0.15.0; upstream is 0.14.2
- package.json — v0.15.0 wins
- CHANGELOG.md — v0.15.0 entry preserved above upstream's v0.14.2
- src/commands/init.ts — upstream wrapped initPGLite + initPostgres in
  try/disconnect blocks and moved engine.initSchema() into the try. My
  aiOpts spreads merged into upstream's refactored config blocks; no
  duplicate saveConfig calls remain.

Build clean: 898 modules, ~160ms compile, 0.15.0 binary runs. 1766 unit
tests pass, 0 regressions.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Locked to 0.17.0 since other PRs (v0.15.x, v0.16.x) may land first.
Also removes the "v0.15" comment in gateway.ts — the v0.15 label belongs
to whatever ships next on master, not this branch.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.15.0 + v0.15.1 + v0.15.2 + v0.15.3 + v0.15.4 + v0.16.0
(llms.txt, hot-issue fixes, bulk-action progress streaming, PgBouncer
prepare:false, durable agent runtime). Test count grows 1766 → 2000+ on
merge. Unit pass 2000/179skip/3 flaky timeouts; the flakes are
pre-existing shared-state setup-hook issues that pass in isolation.

Conflicts resolved:
- VERSION — kept 0.19.0; upstream is 0.16.0
- package.json — v0.19.0 wins
- CHANGELOG.md — v0.19.0 preserved above upstream's v0.16.0/v0.15.x entries
- src/cli.ts — CLI_ONLY merges both: upstream's `agent` + this branch's `providers`
- src/core/embedding.ts — kept gateway delegation but added upstream's
  onBatchComplete progress callback via sub-batching (BATCH_SIZE=100)
- src/core/operations.ts — kept upstream's subagent namespace enforcement
  + kept this branch's gateway.isAvailable() silent-drop fix

Build clean: 939 modules, ~130ms compile, 0.19.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.16.1–v0.18.1: minions worker deploy guide (#287/#317),
subagent Anthropic SDK fix + tsc CI gate (#318), check-resolvable CLI
(#325), dream + runCycle primitive (#321), multi-source brains with
federation + dotfile resolution (#337), RLS hardening + schema backfill
(#343). Test count grows 2000 → 2354.

Conflicts resolved:
- VERSION — kept 0.19.0; upstream is 0.18.1
- package.json — v0.19.0 wins
- CHANGELOG.md — v0.19.0 preserved above upstream's v0.18.1/v0.18.0/v0.17.0/v0.16.x
- src/cli.ts — CLI_ONLY merges `agent`, `providers`, and upstream's new `sources`, `dream`, `check-resolvable`
- src/core/config.ts — merged: kept embedding_model / embedding_dimensions /
  expansion_model / provider_base_urls (mine) + storage (upstream)

Build clean: 948 modules, ~165ms compile, 0.19.0 binary runs. Typecheck green.
18 flaky failures in `bun test` are all PGLite shared-state timeouts in
setup hooks — every failing file passes cleanly in isolation (dream 11/0,
orphans 35/0, check-update 20/0). Pre-existing infra, not introduced by
this merge.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.18.2 (#356): migration hardening + integrity fix +
reserved-connection primitive. New withReservedConnection() method on
BrainEngine interface auto-merged cleanly into pglite-engine.ts and
postgres-engine.ts.

Conflicts resolved:
- VERSION — kept 0.19.0; upstream is 0.18.2
- package.json — v0.19.0 wins
- CHANGELOG.md — v0.19.0 preserved above upstream's v0.18.2

Build clean: 948 modules, ~165ms compile, 0.19.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.19.0 (#326): check-resolvable OpenClaw skills-dir
fallback + docs/tests. Upstream skipped v0.20 to leave room for this
branch at v0.21.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.19.0
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.19.0
- src/cli.ts — CLI_ONLY merges both: my `providers` + upstream's new
  `skillpack`, `routing-eval`, `skillify`

Build clean: 948+ modules, 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.19.1 (#369): smoke-test skillpack (post-restart health
+ auto-fix). New `smoke-test` command added to CLI_ONLY.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.19.1
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.19.1
- src/cli.ts — CLI_ONLY merged: upstream's `smoke-test` + this branch's `providers`

Build clean: 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.20.0 (#195): extract BrainBench to sibling
gbrain-evals repo. Evals move out of gbrain proper.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.20.0
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.20.0

Build clean: 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.20.2 (#364): gbrain jobs supervisor — self-healing
worker process manager.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.20.2
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.20.2

Build clean: 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.20.3 (#379): queue resilience — wall-clock timeouts,
backpressure, --no-worker, env concurrency.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.20.3
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.20.3

Build clean: 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Pulls upstream v0.20.4 (#381): merge gbrain-jobs into minion-orchestrator —
single unified minions skill.

Conflicts resolved:
- VERSION — kept 0.21.0; upstream is 0.20.4
- package.json — v0.21.0 wins
- CHANGELOG.md — v0.21.0 preserved above upstream's v0.20.4

Build clean: 0.21.0 binary runs.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

…providers

# Conflicts:
#	CHANGELOG.md
#	bun.lock
#	package.json
#	src/cli.ts
#	src/core/embedding.ts
#	src/core/import-file.ts

…providers

# Conflicts:
#	CHANGELOG.md
#	VERSION
#	bun.lock
#	package.json
#	src/cli.ts

…providers

# Conflicts:
#	CHANGELOG.md
#	VERSION
#	package.json

Foundation for multi-provider Minions. Purely additive — no behavior change
to existing embedding/expansion paths or to subagent.ts.

- types.ts: 'chat' added to TouchpointKind. New ChatTouchpoint shape with
  supports_subagent_loop separate from supports_tools (Codex F-OV-2: some
  chat-capable models are bad at durable tool loops). supports_prompt_cache
  gates Anthropic-specific cacheControl. AIGatewayConfig gains chat_model
  + chat_fallback_chain.
- Recipe.aliases?: Record<string,string> (Codex F-OV-5). Friendly undated
  forms like 'anthropic:claude-sonnet-4-6' resolve to the dated canonical
  at parse time.
- recipes/anthropic.ts, openai.ts, google.ts: each gains a chat touchpoint.
  Only Anthropic claims supports_prompt_cache=true.
- recipes/deepseek.ts, groq.ts, together.ts: NEW openai-compat recipes.
  DeepSeek powers refusal-fallback + cheap-research. Groq is the speed
  tier. Together is the open-weights house (Qwen, Llama-3.3-70B-Turbo).
- gateway.ts: chat() function wraps Vercel AI SDK's generateText. Returns
  a provider-neutral ChatResult with normalized usage (input/output +
  cache_read/cache_creation pulled from providerMetadata.anthropic per
  D7 review decision). cacheSystem: ephemeral marker only when
  recipe.supports_prompt_cache===true. Stop-reason mapping is
  structural-signal-first per D8 (Anthropic stop_reason='refusal',
  OpenAI finish_reason='content_filter') — refusal regex layer ships
  in commit 3.
- config.ts: GBrainConfig adds chat_model + chat_fallback_chain. Env
  overrides GBRAIN_CHAT_MODEL + GBRAIN_CHAT_FALLBACK_CHAIN.
- cli.ts: connectEngine plumbs chat config into configureGateway.
- providers.ts: --touchpoint chat smoke harness. List shows EMBED/EXPAND/
  CHAT columns. Explain matrix surfaces chat options with input/output
  cost. Recipe alias forms accepted in --model.
- init.ts: --chat-model PROVIDER:MODEL flag.
- test/ai/gateway-chat.test.ts: 21 cases covering recipe registry,
  resolver alias resolution, config plumbing, isAvailable('chat')
  semantics for chat-only/embedding-only providers.

49/49 ai/* tests pass. Typecheck clean.

D11 cross-model resolution. Codex F-OV-1 noted that subagent_messages and
subagent_tool_executions store Anthropic-shaped tool_use / tool_result
blocks as JSONB. When a worker resumes mid-loop and the live model is
OpenAI/DeepSeek, the persisted shape becomes the runtime contract —
read-side translation is lossy.

Mechanical schema-only migration. No code uses these columns yet; commit 2
(subagent refactor onto gateway.chat()) starts writing schema_version=2
with provider-neutral ChatBlock[] in content_blocks.

- migrate.ts: v34 ALTERs subagent_messages + subagent_tool_executions to
  add schema_version (DEFAULT 1) and provider_id (TEXT). All ALTERs use
  ADD COLUMN IF NOT EXISTS so re-runs are idempotent.
- src/schema.sql + pglite-schema.ts: fresh-install DDL gains the same
  columns. New idx_subagent_messages_provider for cost rollups + per-
  provider replay diagnostics.
- schema-embedded.ts: regenerated via bun run build:schema.
- test/migrate.test.ts: 7 new cases pin the migration shape — column
  names + types, idempotency, fresh-install schema parity, embedded
  schema parity. 75/75 migrate tests pass.

Existing rows backfill to schema_version=1 via DEFAULT, tagging them as
legacy Anthropic shape. Subagent.ts read path (commit 2) checks the
version and dispatches the right block mapper.

…providers

# Conflicts:
#	CHANGELOG.md
#	VERSION
#	package.json
#	src/core/migrate.ts

CI's check:privacy gate caught a banned name in src/core/ai/recipes/deepseek.ts:5.
CLAUDE.md (per the privacy rule) bans the private OpenClaw fork name in any
checked-in code. Replaces it with neutral language describing the same
capability ("second hop in a refusal-fallback chain and cheap-research
delegation").

bun run verify now passes locally.

…providers

# Conflicts:
#	CHANGELOG.md
#	VERSION
#	package.json

…providers

# Conflicts:
#	CHANGELOG.md
#	VERSION
#	package.json
#	src/core/migrate.ts