Adds a provider row above Summary (dropdown mirroring Settings' AI
options + a local/hosted badge sourced from llm_status) so it's always
visible which provider a summary/tag generation will actually use.
Generate/Regenerate and the quick switch itself route hosted selections
(Anthropic, or "custom" once its endpoint resolves off-network) through
the same HostedAiBanner one-time acknowledgment gate as Settings — the
generate-click gate is the one that actually matters, since selecting a
provider alone sends nothing; it's kept even though the quick-switch
already pre-empts Anthropic specifically.
Also drops a stale svelte-ignore comment (pre-existing, unrelated to
this change — eslint-plugin-svelte no longer flags that element) that
was failing `eslint .` for this file.
apply_llm_provider_args now resets llm_endpoint to Ollama's local
default when switching to a non-Anthropic provider without an explicit
endpoint and the stored endpoint is still Anthropic's fixed hosted
URL. Prevents a future per-use provider switch (SummaryPanel) that
only sends `provider` from silently leaving llm_endpoint pointed at a
third party for a provider that has no business talking to it.
Adds "Anthropic (Claude)" to the provider select, a write-only masked
API key field (show/hide toggle) shown in place of the endpoint field
for that provider (Anthropic's endpoint is fixed server-side), and a
persistent "leaves this device" note.
Save is gated behind HostedAiBanner the first time the user selects a
hosted provider (Anthropic, or "custom" once its endpoint resolves
off-network) and hasn't acknowledged hosted_ai_acknowledged yet;
accepting persists the ack and completes the save in one step.
New component mirroring ConsentNotice.svelte's card/overlay pattern:
shown before first use of a hosted (non-local) provider, explains that
the transcript leaves the device to a third party, and offers
accept/cancel. Uses the existing design tokens (--accent, --warning,
--bg-elevated, --radius-lg/sm, --shadow-lg) so it matches the
recording-consent notice visually. Not yet wired into any view — next
commits add it to Settings and SummaryPanel.
settings.svelte.ts's setLlmProvider now accepts an optional apiKey
(forwarded straight to the set_llm_provider command, which is the only
place it gets stored — the OS credential store) and DEFAULT_SETTINGS
gains hosted_ai_acknowledged. acknowledgeHostedAi() persists the
one-time hosted-AI banner acknowledgment the same way
acknowledgeConsent() does for recording consent.
set_llm_provider now accepts provider "anthropic": the apiKey argument
is written straight to crate::llm::credentials (OS credential store)
and never assigned into Settings, so it structurally cannot reach
settings.json/wa.db (FR-SEC-1). apply_llm_provider_args pulls the
settings-mutation logic into a small pure function specifically so
that guarantee is unit-testable without touching a real keyring.
llm_provider_from_settings gets an "anthropic" arm; api.anthropic.com
joins the settings-derived egress allowlist through the exact same
generic is_local()-based path privacy_self_check_json already uses for
every other provider — no anthropic-specific allowlist code, and the
host only appears once the provider is actually selected (never
unconditionally).
Also fixes pre-existing cargo-fmt drift on one unrelated line this
file's formatter pass touched (CONTENT_RANGE header call).
One-time "data leaves your device" acknowledgment flag for hosted
(non-local) AI providers, persisted like consent_acknowledged so the
banner doesn't nag on every use. serde(default) so it's false for any
settings.json written before this field existed.
Implements AnthropicProvider::status/summarize/suggest_tags against the
real Anthropic Messages API (POST /v1/messages, x-api-key + anthropic-
version headers, SSE streaming for summarize, non-streaming for the
short tag reply), replacing the stub that returned "isn't built yet".
- Adds llm::credentials (mirrors sync::credentials) so hosted API keys
live in the OS credential store, never settings/DB/logs (ADR-0011).
- Fixes OpenAiCompatProvider::api_key() to actually read credential_ref
from the store instead of the hardcoded None left by the T10a.2 stub.
- Splits build_messages/build_tag_messages into system/user halves
(build_system_and_user/tag_system_and_user) since Anthropic's system
prompt is a top-level field, not a messages[0] entry like OpenAI's shape.
- is_local() is unconditionally false for AnthropicProvider (no
unauthenticated mode, unlike OpenAiCompatProvider).
- Adds a loopback raw-socket mock HTTP server (mirrors sync::oauth's
LoopbackRedirect pattern — no HTTP-mock crate in the dependency tree)
and tests proving request/response shape for both AnthropicProvider
and OpenAiCompatProvider against mocked endpoints.
Written to satisfy "an in-process MCP client can call the tools", but
its dependency (reqwest 0.13 via rmcp's client-http feature) pulled in
a very heavy native build (aws-lc-rs) that hadn't finished verifying
when told to stop building. Removing rather than leaving an unverified
test + dependency in the tree. The acceptance criterion it targeted is
still exercised at the unit level: http_transport::tests (loopback
bind/refusal) and mcp::scope::tests (tool-handler scope logic) both
pass under --features mcp; the full wire-protocol round trip is a
follow-up (see final report).
Per instruction to stop iterating on builds: reqwest 0.13's client-http
feature (needed only for an in-process MCP-client e2e test) was pulling
a very slow/heavy native build (aws-lc-rs, even with the TLS backend
stripped down) and the last attempt was still in progress -- unverified.
Reverting Cargo.toml to the last state that was actually confirmed
green (cargo fmt/clippy -D warnings/test, both default and --features
mcp, all passing) rather than ship an unverified dependency change.
The HTTP transport itself (bind_loopback/token gate/serve loop) keeps
its own unit tests, which did pass under --features mcp.
The e2e test only ever talks plain http://127.0.0.1 (never https://), so
a TLS backend is unnecessary weight -- reqwest 0.13's `rustls` feature
pulls in aws-lc-rs, a large C codebase that was taking a very long time
to compile (and once contributed to a pagefile-exhaustion build failure
alongside concurrent cargo invocations). Building reqwest013 with
default-features=false and no TLS feature is enough for the plain-HTTP
client the test needs.
rmcp's transport-streamable-http-client-reqwest is written against
reqwest 0.13.2+, a semver-incompatible major vs. the reqwest 0.12 the
rest of WA (llm/sync) depends on -- Cargo can't unify those, so
`reqwest::Client` from our own dependency isn't the same type rmcp's
`impl StreamableHttpClient for reqwest::Client` is written for. Adds
`reqwest013` (a Cargo package-rename of reqwest 0.13) purely so
tests/mcp_server_test.rs can construct the exact client type rmcp
expects, without touching reqwest 0.12 anywhere else in the crate.
A real rmcp client speaks Streamable HTTP to the real RmcpServer stack
(bind_loopback + serve + WaMcpHandler) over an OS-assigned loopback
port, using an in-memory SqliteStore -- not just isolated unit pieces.
Asserts: an unauthenticated POST is rejected (401) before it ever
reaches the MCP service; an authenticated client's list_tools/call_tool
round-trip is schema-valid; and the call appends an mcp_access_log row
even though default scope (`none`) denies it data (FR-MCP-5 logs the
ask, not just the answer). mint_and_store/delete touch the real OS
credential store, same pattern as sync::tests::webdav_round_trip.
Widens visibility from pub(crate) to pub so tests/mcp_server_test.rs
(an external integration-test crate, per CLAUDE.md's "cross-service
tests in /tests") can drive the real HTTP transport end-to-end. This
does not weaken the loopback guarantee: bind_loopback still refuses
non-loopback regardless of caller -- only its visibility changed.
New "MCP server" tab: persistent disclosure banner (agents may forward
served data to their own provider; WA itself adds no egress), enable
toggle, transport/port/scope controls, a reveal-once auth-token callout
(aria-live, never re-shown), and a live access-log list backed by the
settings store's mcp:// access subscription (FR-MCP-5).
Privacy tab gains a compact MCP status row + an explicit "adds nothing
to the egress list" confirmation line (FR-MCP-7, FR-SEC-2) next to the
existing sync/LLM egress rows.
Built with ui-ux-pro-max guidance (disclosure banner wording, reveal-
once secret pattern, aria-live for the token, badge semantics) mapped
onto this file's existing minimal/utilitarian style (banner/badge/row/
confirm classes already used by the Sync and Privacy sections) rather
than introducing a new visual language.
loadMcpStatus/loadMcpAccessLog/setMcpEnabled/setMcpScope mirror the
existing sync-target patterns; a live "mcp://access" subscription tails
the FR-MCP-5 audit log into the store while the panel is open, on top
of the on-demand loadMcpAccessLog() refresh. mcpLastToken holds the
freshly-minted token from the most recent enable, for the one-time
reveal UI (next commit).
Adds mcp_transport/mcp_port/mcp_expose/mcp_expose_recordings to
AppSettings and a typed McpStatus for mcp_status()'s response; types
onMcpAccess's payload correctly (the live event is camelCase per
docs/04-api-contracts.md, unlike the snake_case McpAccessEntry rows
mcpAccessLog() returns).
Regenerated by `cargo check --features mcp` after promoting hyper/
hyper-util/http/http-body-util/bytes/tower-service/tokio-util to direct
deps and widening rmcp's feature set (see the earlier Cargo.toml
commits).
iamdoubz