* fix(chat): warn before "regenerate" permanently deletes later messages
Both "regenerate" entry points — the per-message footer's ↻ "Regenerate
from here" button, and the vision editor's "Regenerate message" button —
call POST /api/session/{id}/truncate, which does a real, permanent SQL
delete of every message after the point clicked, then generates a new
reply. Clicking either on anything but the very last exchange did this with
zero warning: the button's own label doesn't say it also erases the rest
of the conversation, and neither button visually signals it's destructive
differently from any other message-footer action.
Caught live: a mis-click destroyed a real conversation with no way to
recover it (no backup covered the affected window). This adds a
confirmation prompt — via the existing uiModule.styledConfirm — whenever
the action would delete more than just the message(s) being regenerated,
stating the exact count of messages that would be lost. Regenerating the
most recent exchange (the everyday case, nothing exists after it) is
unaffected and still proceeds with zero friction.
New test: tests/test_regenerate_truncate_warning_js.py — drives the real
regenerateFrom/resendUserMessage functions under Node with a fake message
list DOM, confirming: the last-exchange case never prompts and truncates
directly; an earlier case prompts with the correct count; cancelling
aborts before any network call; confirming proceeds with the correct
keep_count. Also fixed a pre-existing, unrelated gap this surfaced while
verifying the full suite: test_pr6020_browser_review_regressions.py's
Node harness never stubbed `navigator` at all (sessions.js reads
navigator.platform at module load) — a gap on origin/dev itself, not
caused by this branch's diff, using the same Object.defineProperty fix
already applied on #6019/#6148 for the sibling issue there (Node 21+'s own
getter-only navigator global, not relevant to this specific failure but
the same defensive pattern).
Verified: full pytest suite in-container, 5798 passed / 11 pre-existing
unrelated failures (README/docs/agent-loop baseline) / 0 new.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
* test(chat): cover plain resend after dev's replace-by-default change
dev made plain resend (no options) replace from the clicked turn by
default, with append-only resend opt-in via { append: true } (see
test_resend_message_nondestructive.py). The footer "Resend message"
button calls resendUserMessage(el) with no options, so it now truncates
everything after the clicked turn and goes through this PR's warning.
Replace the old "plain resend never prompts" test, which assumed the
previous append-by-default behavior, with one asserting that plain
resend on an earlier pair prompts with the right count, and one
asserting that explicit append-only resend never truncates or prompts.
Also resolve ui.js/sessions.js through the same cache-busting query
string chat.js imports them with. ES modules cache by full URL, so the
harness's unversioned import of ui.js was a second module instance and
its uiMod.styledConfirm stub never reached the instance chat.js calls.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(chat): warn before an edit deletes later messages; offer to fork instead
Sending an edited user message truncates the chat at that message, the
same unrecoverable server-side delete as regenerate and resend, but did
it with no prompt. It now goes through the same confirmation, worded for
the action ("Sending this edit will also permanently delete the N
messages after it"). Cancelling leaves the editor open with the draft.
The confirmation for all three actions (regenerate, resend, edit) also
offers "Fork from here": it forks the chat just before the user turn,
switches to the new chat, and replays the action there (the same message,
or the edited text), leaving the original chat untouched. forkFrom now
shares the same fork helper.
Tests drive the real editUserMessage Send handler under Node and cover
the edit prompt (count, cancel, last-message no-prompt) and the fork
choice for each action (fork keep_count, no truncate, replayed text).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(chat): cut and fork a chat by message id, not loaded-page position
Regenerate and edit sent keep_count, the clicked message's position among
the messages loaded in the page. A long chat loads only its newest messages
until the user scrolls up, so that position is smaller than the real one and
the truncate deleted real messages above the clicked one (regenerating the
latest reply in a ~50-message chat removed about a dozen earlier messages).
Resend already sent before_msg_id; regenerate and edit now do the same.
The fork options this PR adds had the same flaw. Fork now accepts
before_msg_id (fork before that message) and through_msg_id (include it),
resolved against the stored messages; keep_count stays as the fallback for a
message that has no id yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(chat): keep a regenerated turn's message id current; stop on a refused cut
Regenerate keeps the original user bubble on screen and resubmits its text
without drawing a new bubble. The resubmitted message is a new DB row, but
the bubble kept the deleted row's id, so the next regenerate's cut was
refused (404) and the page sent the turn anyway, stacking a second reply.
The stream now reports the saved user message's id (user_message_saved),
which the page attaches to the user bubble, or to the bubble a regenerate
kept. Regenerate and edit stop with an error when the server refuses the
cut instead of sending.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(chat): name the buttons in words instead of symbols
CONTRIBUTING asks for no Unicode emoji in UI or code; the test comments
used the regenerate and pencil glyphs to name buttons.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* refactor(chat): share the fork-and-send step across edit, resend and regenerate
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(chat): pass the regenerate fork its attachment ids explicitly
Resend already hands its collected file ids to _forkAndSend; regenerate
relied on the earlier _pendingRegenAttachments assignment. Pass _regenIds
too so both fork paths read the same, and check the hand-off in the
regenerate tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(chat): assert fork choices keep the user message's attachments
Run regenerate and resend's "Fork from here" against a user message with
an uploaded file and check that the send fired in the fork carries its
file id, replacing the source-text check. chat.js exposes a read-only
_pendingRegenAttachmentIds() for this, like cookbook-hwfit.js's test hooks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(chat): preserve attachments across edit and regeneration
---------
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
Co-authored-by: Alexandre Teixeira <general@alexandreteixeira.dev>
* feat: add Notes and Todos as agent chat context
- Send selected Note/Todo IDs from the Notes UI
- Load selected notes with owner scoping in chat routes
- Format note content and checklist items for the model
- Inject selected notes as untrusted chat context
- Add a selected-context indicator beside Shell Access
- Keep selected context associated with the active chat session
* Improved selected notes context handling and discouragin redundant note lookup and permission requests
* Fix selected notes UI when switching chats
* fix(notes): harden selected context lifecycle and validation
---------
Co-authored-by: Alexandre Teixeira <111787685+alteixeira20@users.noreply.github.com>
Chat/agent harness work: request-authority model, turn contract, tool
policy/parsing/execution, clean_agent_preview streaming/thinking, ask_user
persistence, session probes, and scheduler.
Source: vendor/photocraft (ad1cff44) harness files, 3-way merged onto
upstream/dev (a8c147b2). Harness core files were independently developed
in both lineages; resolution policy documented in pr_work/C_conflicts.md
(vendor kept where ahead, upstream kept where vendor is behind; full delta
preserved in pr_work/C_upstream_delta.patch).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Narrowly guard _request_privileges() in routes/chat_routes.py against
synthetic requests lacking scope['app'] or auth manager state, safely
returning empty privileges without granting agent privileges.
- Add focused regression test in tests/test_context_resolution_route.py
verifying that requests without app scope do not crash and cannot gain
agent privileges or qualify for compact preview runtime.
- Regenerate website/configuration-reference.md mechanically to align with
current source line numbers.
The chat route repeated the compact (clean v3) eligibility decision inline
to prepare the turn's context resolution, while the agent loop dispatched
on the contract stamp set by a separate, later condition. The two could
drift, and already disagreed for a user whose privileges demote the turn
to plain chat: the route prepared a compact resolution that no compact
runtime used.
src/agent_runtime/runtime_selection.py (no imports) now owns the rule:
- uses_compact_preview_runtime(): clean route requested, contract policy
enabled, agent mode, agent permitted, not an image generation session.
- is_compact_preview_contract() and COMPACT_PREVIEW_MODE for the stamp.
The route evaluates the rule once, before context preparation, where all
of its facts are final (the agent privilege is read through the same
_request_privileges helper the later enforcement uses). That one value
gates the typed context resolution and is the _clean_v3_preview flag that
stamps the contract; inside the agent-contract branch it equals the
previous condition, so stamping behavior is unchanged. The agent loop
dispatches through is_compact_preview_contract(), and the compact runtime's
MODE is the shared constant.
A route-level matrix drives the real agent loop and asserts that route
preparation and compact dispatch agree for compact, escalated, configured
compact/full, regular, TUI, privilege-denied and image-generation turns.
The first checkpoint removed terminal-metrics discovery, but a normal
compact chat turn still ran two context systems: build_chat_context's
legacy untyped lookup (directly or inside maybe_compact) and the typed
resolver inside stream_preview.
Resolve the typed ContextResolution once, at the chat route, before
build_chat_context, using the session's provider credentials. The
predicate mirrors _clean_v3_preview; every input it needs is known at
that point and the native-workspace term cannot veto a requested clean
route. The same object then:
- sizes legacy history shaping in build_chat_context through a new
maybe_compact(context_length=...) override, so no legacy probe runs;
an unknown window still shapes with DEFAULT_CONTEXT but gains no
provenance;
- crosses stream_agent_loop (one new parameter, forwarded only at the
compact dispatch) into stream_preview, which reuses it and probes only
for callers that arrive without one or with one bound to another
route.
ContextResolution now records the endpoint and model it describes
(endpoint URL excluded from repr and metrics). The bare legacy
context_length is never converted into typed evidence.
Credential scoping: origins compare with default ports normalized, an
empty host is never trusted, and the probe client never follows
redirects. Tests cover the configured origin, the server-resolved
Tailscale form, scheme/port/lookalike/userinfo/path origins, redirects,
and secret-free errors, logs and metrics.
The conftest guard now replaces only the resolver's I/O edges (HTTP
client and DNS-capable URL building) instead of the whole probe, and
exposes a context_probe_ledger fixture, so route integration tests run
the real resolver offline and can count metadata requests.
Snapshot current maintainer-preview application changes and regression fixtures for integration into lab. Excludes local runtime data, evaluation outputs and source backups. Focused Python regression selection: 140 passed; full suite not certified.
* fix(security): stop API tokens reaching privileged agent tools
A bearer API token resolves to the human who minted it, and minting is admin-only, so every owner-keyed privilege check in the agent path answers "admin". A token issued for a narrow integration therefore reached bash and python with the authority of the account that created it.
Three independent routes to that sink, each closed here.
The token could answer its own tool-approval prompt. An approval records that a person authorized one dangerous action, and a token cannot make that statement, so /api/chat_stream now refuses an approval resume from a bearer caller.
The chat-session grant was reconstructable from caller-supplied message metadata. Two routes persist a metadata blob on the caller's behalf, so the shape of a resolved approval card could be written straight into a transcript and was then read back as authority. The server now signs the grant when it resolves an approval and verifies that signature when reading it back, binding it to the chat and the approval it was issued for. Both routes also drop server-owned keys from an inbound blob.
A run driven by a token inherited its owner's tool set. Such a run is now capped at the non-admin policy regardless of who minted the credential, which holds even where no approval is raised at all.
The human path is unchanged: a browser session still receives the prompt, still approves, and a granted chat-session scope still carries to later turns in that chat.
Scope enforcement across the wider route surface is a separate gap and is not addressed here.
* fix scoped chat delegation boundaries
* fix(auth): reject malformed chat approval signatures
---------
Co-authored-by: RaresKeY <158580472+RaresKeY@users.noreply.github.com>
* fix(agent): allow remaining actions for an approved task
* fix(agent): make approval continuation control-only
* fix(ci): preserve approval taint and cache-buster contract
* fix(ui): keep tool approvals in current chat
* fix(ui): route tool approvals through chat submit
* test(ui): pin approval submit routing
* fix(agent): complete approval denial flow
* fix(ui): avoid duplicate ask-user close icon
* fix(agent): retain approved tool in continuation set
* revert(ui): keep PR 6113 scoped to approval continuation
* fix(agent): add task and chat approval scopes
* fix(ui): prevent duplicate ask-user close icon
* feat(ui): add ask-user option shortcuts
* fix(compare): route ask-user choices per pane
* fix(agent): keep skill-test approvals to a single action
The chat card now reuses the wire value `approve` to mean chat-session
scope, and `consume()` returned `allow_remaining_actions=True` for it
unconditionally. The skill-test approval route was never updated: it still
sends `approve` meaning "once", and its button still reads "Allow once",
but the grant it got back set `approval_gate_bypassed` for the rest of the
resumed run. That surface wraps the skill body and every transcript byte
as untrusted context, so it is the last place where one click should
ungate everything that follows.
Give `consume()` an explicit `allow_continuation` flag. Callers that own a
resumable chat keep the scope the user picked; callers that do not — the
skill tester, unattended audits — get SINGLE_ACTION and the gate re-arms
behind the sealed action, which is what their label promises.
* fix(ui): cache-bust every module the approval click depends on
chatStream.js, compare/index.js and compare/stream.js all changed
behaviour but kept their old `?v=`, while chat.js and chatRenderer.js were
bumped. A returning browser therefore serves the new chat.js — which now
deliberately leaves the composer empty and clicks the send button — next to
the cached chatStream.js that has no interceptor. With an empty composer
that button sits at `data-mode="newchat"`, so the click opens a new chat
and the approval is dropped.
Bump the three, and version compare/stream.js's chatRenderer import to
match everyone else's so the ask_user keydown listener binds to one module
instance instead of two.
* fix(ui): keep the digit shortcuts off tool approval cards
With an approval card on screen and focus anywhere outside an input, a bare
`1` fired `approve_task` — the widest of the three grants — with no
modifier and no confirmation. That card is the one control whose entire
purpose is deliberate consent after untrusted context influenced the run,
and Deny sits at 3.
Label the card with its kind and skip the shortcut for approvals. Ordinary
ask_user questions keep 1-3.
* fix(compare): restore a pane's ask_user card instead of dropping the choice
renderAskUserCard removes the card as soon as onSubmit accepts, but the
resume loop gave up silently after 10s if the originating stream still owned
the pane. The user saw the click land, the card vanish, and nothing happen,
with no way to get it back.
Re-render the card on that deadline and say why. The reroll case still
returns without sending — that choice belongs to a stream that no longer
exists.
* refactor(chat): drop the unreachable deny branch
`if decision != "deny"` is always true — the deny path returns a
StreamingResponse a few lines above. It reads as if deny still falls
through to the toggle restore.
---------
Co-authored-by: Léo <leograndcontact@gmail.com>
* refactor(model-routing): centralize explicit foreground fallback policy
Make foreground fallback an explicit per-user, availability-only policy shared by streaming Chat, non-stream Chat, and Agent runs.
Preserve strict defaults, owner/model and credential boundaries, pinned Agent routes, and truthful per-round provenance/accounting. Carry provider-reported model identifiers through native streaming adapters, non-stream responses, and caches, and keep legacy default_model_fallbacks as tombstoned raw storage that generic settings APIs and agent tools cannot expose or mutate.
* fix(agent-loop): restore rebase-dropped qwen routing, workspace prompt, and temperature clamp
* fix(model-routing): thread selected endpoint identity, fix cost classification and fallback eligibility
* fix(chat): restore stream helpers and harden run stop lifecycle
* fix(model-routing): let numeric provider codes win over symbolic rate-limit statuses
* fix(agent-loop): apply qwen temperature and notes-tool clamps per fallback candidate
* fix(chat): honor queued stop across resend and reload canonical terminal on EOF
* fix(chat): track stop queue and cleanup ownership by per-send generation
* fix(agent-loop): preserve requested temperature for non-qwen fallback candidates
* fix(chat): reserve send ownership before any await and scope stop to the current send
* fix(chat): clear the previous run identity at send reservation
---------
Co-authored-by: RaresKeY <158580472+RaresKeY@users.noreply.github.com>
Co-authored-by: StressTestor <212606152+StressTestor@users.noreply.github.com>
chat_stream() references `_explicit_web_intent` in three places
(disabled-tools gating, global-disabled web allowance, and the
per-turn tool filter) but the assignment was dropped during a
branch merge. Every chat request raised
NameError: name '_explicit_web_intent' is not defined
at routes/chat_routes.py, surfacing to the client as a bare
"Internal Server Error" before any LLM call was made — chat was
fully broken on dev and main.
Restore the original definition, computed from the already-derived
tool intent, immediately before its first use:
_explicit_web_intent = bool(_tool_intent and _tool_intent.category == "web")
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Guard the agent_max_tool_calls settings read so hand-edited or agent-written non-numeric settings.json values fall back to 0 instead of crashing agent-mode chat stream initialization. Add regression coverage for guarded coercion.
* fix(security): redact credential-bearing URLs and PII from logs
Several log statements emitted sensitive data in clear text:
- model_routes / chat_routes / contacts_routes logged endpoint URLs raw.
Admin-configured URLs can embed credentials in userinfo or query
(e.g. https://user:pass@host, ?api_key=...). Route them through a
shared core.log_safety.redact_url() that drops userinfo/query/fragment.
- note_routes / task_scheduler logged operator email addresses (smtp_user,
recipient). Replaced with presence booleans, which keeps the diagnostic
("why didn't this send") without writing PII to logs.
model_routes already had a local redactor on its HTTPStatusError branch;
the generic except branch was missed, so reuse the existing helper there.
Clears CodeQL py/clear-text-logging-sensitive-data alerts 264, 317, 324,
325, 343, 344, 528.
* fix(security): re-bracket IPv6 hosts and single-source the URL redactor
Address review on #4750:
- redact_url now re-brackets IPv6 literals so host:port stays
unambiguous (https://[2001:db8::1]:8443/v1, not the bracket-less
ambiguous form).
- point model_routes._redact_url_for_log at the shared helper so the
two redactors are single-sourced (also picks up the IPv6 fix).
* fix(api): attribute bearer-token actions to the token owner on owner-scoped routes
Owner-scoped chat, session, and upload routes called
get_current_user(), which resolves a bearer ody_ API token to the
sandboxed "api" pseudo-user. A paired API-token client (companion, CLI,
IDE extension) therefore saw and created a separate "api"-owned silo
instead of the owner's data.
effective_user() already exists for exactly this: it attributes a token's
actions to request.state.api_token_owner, is identical to
get_current_user() for cookie sessions, and falls back safely when a
token has no owner. session_routes.py was already migrated; this
completes the migration for the remaining owner-scoped routes:
- chat_helpers.py: chat-privilege enforcement, message attribution, prefs/context
- chat_routes.py: orphaned-endpoint owner, session-auth owner, message search
- upload_routes.py: upload owner attribution + access checks
The /api/models swap is intentionally omitted: #4292 already migrated it
to effective_user (plus the chat-scope gate and ownerless-token 403), so
this PR keeps dev's version of routes/model_routes.py unchanged.
chat_routes.py keeps importing get_current_user for the workspace owner
gate; session_routes.py drops the now-unused import.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test: target effective_user in auth monkeypatches and owner-scope assertion
The owner-scoped routes now call effective_user() instead of
get_current_user(), so the tests that stubbed get_current_user (or
asserted on it) follow suit:
- test_chat_helpers.py, test_review_regressions.py,
test_kv_cache_invalidation_2927.py: monkeypatch effective_user
- test_session_endpoint_owner_scope.py: assert the owner-scope guard uses
effective_user(request)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
