The claude-code provider spawns the bundled `claude` CLI via
@cline/sdk → ai-sdk-provider-claude-code → @anthropic-ai/claude-agent-sdk.
On macOS, when that subprocess is launched from inside the VS Code extension
host it performs IDE auto-connect, which shells out to `osascript`
(`tell application …`) to identify/activate the editor — raising the macOS
"control other applications" (Apple Events / Automation) permission prompt.
There is no AppleScript/osascript anywhere in the JS dependency tree (the only
match is the build-only `vite`); the Apple Event is sent by the spawned `claude`
binary, which contains 26 osascript calls. Every one is user-action-gated
(clipboard image, screenshot, terminal-setup, Claude-Desktop deep link, browser
detection) except the IDE auto-connect path, which is the only one that fires
without user action when launched inside an IDE — exactly the reported symptom
(appears on use, harmless when declined, since we never use the connection).
Fix: spawn the agent with the macOS IDE integration disabled via the provider's
`options.env` passthrough — `CLAUDE_CODE_AUTO_CONNECT_IDE=0` hard-returns out of
the binary's auto-connect gate (`mK("0") === true`), and
`CLAUDE_CODE_IDE_SKIP_AUTO_INSTALL=1` skips the extension-install deep link.
`options.env` is seeded with the full `process.env` so the default environment
inheritance is unchanged (login under HOME, proxy/CA vars preserved) — only the
two switches are added. The env flows end to end into the spawned child
(verified through the provider's getBaseProcessEnv merge and the agent SDK's
spawn). runEditTurn is the sole agent-spawn site, so this covers every flow.
Unit test asserts the agent is constructed with the two switches set and the
rest of the environment preserved. 250 unit + typecheck + build green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Surface Claude live output/progress during the asking-Claude status: notification activity line + token count, a Cowriting: Claude OutputChannel streaming assistant text, and a cancellable turn. Pure turnProgress reducer + runEditTurn onProgress/AbortSignal (vscode-free) + both call sites. INV-43..47.
Fixes#60
The F9/F10 author-coloring technique injects paired PUA sentinels into prose
source at span offsets, renders, then maps sentinels → cw-by-* spans. Two failure
modes when a span boundary met emphasis markup (characterized in session 0032):
- CASE1: a boundary strictly inside a delimiter run (a**b**c, between the two *)
split `**`, breaking markdown-it's delimiter pairing (stray <em></em>, raw **).
- CASE3: a boundary inside an emphasis run (**bold** span covering **bo) rendered
the emphasis but MISNESTED span/element (<strong>bo</span>ld</strong>).
Token-aware fix (both, per the issue #33 comment):
- injectSentinels: clamp any sentinel offset that lands strictly inside a
delimiter run (* _ ~ `) to the run's start, so a sentinel never splits a run
(CASE1). Delimiters are invisible once rendered, so this only shifts the colored
boundary across markup. Drop spans that clamp to empty.
- sentinelsToSpans: replace the naive global split/join with a walker over the
rendered HTML that emits the cw-by-* span only around TEXT runs — closing it
before any <tag> and reopening after — so a span is always well-nested within
inline elements (one span segment per text run, CASE3). Tags are copied
verbatim with any stray sentinel stripped (no Private-Use-Area char leaks).
Pure, vscode-free, deterministic (INV-33). No regression: existing colorByAuthor /
renderReview cases stay green; the common no-emphasis case is byte-identical.
222 unit + 74/5 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A pinned baseline with no changes since should read as a clean starting point,
but the F10 on-render still author-colored every block (colorByAuthor), painting
the whole document green/blue right after a pin. Now, when the baseline reason is
"pinned" and there are zero changes since (every diffBlocks op unchanged), the
on-render is fully clean — no change marks and no authorship coloring — while the
data-src block mapping (INV-36) and any pending proposals (review actions, not
annotations) are kept.
Scoped to the PIN specifically, not all zero-diff: a baseline advanced by a
machine-landing (accept) is also zero-diff but keeps its authorship coloring so
accepted Claude text stays blue (F10 INV-33). renderReview takes a `pinned`
option; the controller passes baseline.reason === "pinned" from both refresh and
the renderHtmlFor test seam.
- trackChangesModel.ts: renderReview gains the `pinned` RenderOption; when pinned
+ zero-diff, blocks render plain (no colorByAuthor).
- trackChangesPreview.ts: pass { pinned } through refresh + renderHtmlFor.
- unit: pinned+zero-diff → no cw-by-*; pinned+zero-diff still shows proposals;
zero-diff WITHOUT pin (machine-landing) keeps coloring (INV-33); real changes
after a pin re-color.
- s48PinClean host E2E: type → colored; pin → clean; edit → annotations return.
218 unit + 74/5 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document-edit flow SLICE-3 — completes reach→review→accept
(specs/coauthoring-document-edit-flow.md §7.2, INV-42). A single "Accept all"
gesture applies every pending proposal on the current document through the
existing F4 acceptById seam (block proposals take the INV-40 word-precise path
automatically), in descending anchor order so an earlier accept never invalidates
a later one, skipping (never force-applying) proposals that can't anchor and
reporting applied-vs-skipped. Batched application of the existing accept path — no
new mechanism; the webview posts intent only (INV-35). No confirmation dialog
(undo restores).
- proposalController.ts: acceptAllProposals(document) → {applied, skipped}
(descending order, orphan-skip); accept/acceptById gain a silent opt so the
batch suppresses N per-proposal orphan warnings in favour of one report.
- trackChangesPreview.ts: ToolbarMsg += {type:"acceptAll"}; handleWebviewMessage
routes it to a public acceptAll(document) that batches + reports.
- extension.ts + package.json: cowriting.acceptAllProposals command (active doc,
markdown-gated palette entry) for the non-webview path.
- media/preview.ts + shellHtml: "✓✓ Accept all" toolbar button, posting the
intent, shown only with ≥2 pending proposals (authorable, on-state).
- trackChangesModel.ts: diffToBlockHunks now emits one block-aligned hunk per
CHANGED block even when changed blocks are ADJACENT (treats changed blocks as
1:1 anchors alongside unchanged ones; gap-spans only cover add/remove runs
between anchors) — fixes adjacent changed blocks collapsing into one proposal.
- f12Accept host E2E (apply-all reconstructs; orphan skip + report; single
proposal; command registered/gated); MANUAL-SMOKE-F12 §3.
214 unit + 73/5 host E2E green. Completes the document-edit-flow cluster
(#42 reach + #47 review + #46 accept).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document-edit flow SLICE-2 (specs/coauthoring-document-edit-flow.md §7.2,
INV-39/40/41 — P1 "too much to review"). A whole-document rewrite now proposes
ONE F4 proposal per CHANGED BLOCK (the unit a human reviews), but accepting a
block reconciles attribution at WORD granularity (the unit F3 records). Block =
decision unit; word = attribution unit. Supersedes INV-37's per-word cut for
document edits; selection edits unchanged.
- trackChangesModel.ts: new pure diffToBlockHunks(current, rewritten) — block-key
alignment (reusing diffArrays/diffBlocks keying): an isolated changed block →
one block-aligned hunk → the rewritten block raw (a code/mermaid fence is one
atomic whole-fence hunk, INV-23); insert/delete runs → one gap-span hunk over
the inter-anchor region (separators included) so reconstruction stays exact; a
zero-width gap-span is anchored (INV-41). Also split diffToHunks into the raw,
un-anchored wordEditHunks + the anchoring wrapper (the anchoring could grow an
insertion to overlap an adjacent hunk, corrupting a batch apply — a latent bug
that only surfaced once hunks are applied as a batch).
- trackChangesPreview.ts: runEditAndPropose document branch uses diffToBlockHunks
and tags each proposal granularity:"block".
- model.ts / proposalModel.ts: additive optional Proposal.granularity
("block"|"single"; absent ⇒ single, back-compat — no migration).
- proposalController.ts: accept of a block proposal runs an intra-block word
sub-diff (wordEditHunks, disjoint) and applies one applyAgentEdit per changed
run, descending offset — only the words Claude changed land Claude-attributed;
unchanged spans keep prior authorship (INV-40).
- Tests: diffToBlockHunks unit (reconstruction + fence atomic + add/remove);
f12Review host E2E (M blocks→M proposals, unchanged→none, fence atomic, INV-40
attribution, INV-41 insertion accept); updated the f11 document-path E2E to
per-block (INV-39 supersedes INV-37); MANUAL-SMOKE-F12 §2.
Seam note: pendingEdits.matchEvent resolves one registration per change event, so
INV-40's per-run attribution is sequential applyAgentEdit calls (N undo steps),
not one multi-replace WorkspaceEdit — see transcript Deferred decisions.
214 unit + 69/5 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Document-edit flow SLICE-1 (specs/coauthoring-document-edit-flow.md §7.2,
INV-38): make "Ask Claude to Edit" reachable from the editor body AND the
editor tab, selection-aware — a selection routes to editSelection, no
selection to editDocument — both gated to markdown/authorable docs, both
flowing through the single runEditAndPropose path (no divergent edit code).
- package.json: add editSelection + editDocument to editor/context (selection-
aware, markdown + file/untitled) and editor/title/context (selection-aware,
resourceLangId == markdown). Markdown-gate the existing editor/context
editSelection entry to match (its command handler is unchanged; the palette
still reaches any authorable doc — see transcript Deferred decisions).
- trackChangesPreview.ts: cowriting.editDocument accepts the clicked tab's
resource Uri (editor/title/context), targeting THAT document (opening it if
needed) and falling back to the active editor when invoked with no arg —
mirroring showTrackChangesPreview's #41 clicked-doc resolution.
- E2E (test/e2e/suite/f12Reach.test.ts): menu entries present, selection-aware,
markdown-gated; editDocument(uri) targets the tab doc not the active editor;
no-arg falls back to the active editor.
- docs/MANUAL-SMOKE-F12.md: SLICE-1 reach smoke steps.
208 unit + 65/5 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The review preview (`cowriting.showTrackChangesPreview`) was only reachable from
the command palette and `ctrl+alt+r` — neither where a writer's hand naturally
goes. Add the obvious right-click entry points to the plugin's central surface:
- `explorer/context` + `editor/title/context` menu items gated on
`resourceLangId == markdown`, both invoking `showTrackChangesPreview`.
- The command now resolves the *clicked* document: it opens the passed Uri if
it isn't already an open document (the Explorer case), instead of falling back
to the active editor. No-arg invocation (palette / keybinding) is unchanged.
- Retitle the command to "Open Cowriting Review Panel" so the menus read the
operator's wording (palette shows "Cowriting: Open Cowriting Review Panel").
E2E: clicked-doc resolution (open + not-yet-open + no-arg fallback), both menu
contributions present + markdown-gated, title, and keybinding unchanged.
Closes#41.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Code-review follow-up on the F11 branch (5966907..1564ef5). Three real findings:
- CRITICAL — pure-insertion hunks were born-orphaned. A document rewrite that
INSERTS text produced a zero-width hunk (start==end) → buildFingerprint yields
empty fp.text → anchorer.resolve orphans an empty needle → the proposal could
never be accepted (silently). diffToHunks now anchors every zero-width
insertion to an adjacent source token (anchorInsertion): the range absorbs the
token and the replacement keeps it, so net text is identical but fp.text is
non-empty and resolvable. New unit tests: applying hunks always reconstructs
the rewrite (substitute/delete/insert/multi); insertions are never zero-width.
New host E2E ACCEPTS a rewrite-with-insertion end to end and asserts accept-all
reconstructs the intended document (also covers sequential multi-hunk accept).
- IMPORTANT — renderPlain cross-block fidelity. The per-block off-mode rendering
(SLICE-2) can't resolve a reference-link definition in a separate block.
Documented as a conscious tradeoff of the operator-locked block-level mapping
(§6.7) — renderReview already rendered per-block, so this keeps both modes
consistent — with a characterization test + a docstring note.
- MINOR — F11 proposals now carry a turnId (one per Ask-Claude gesture, shared
across a document rewrite's N hunks), matching the editor-menu editSelection
path so a rewrite groups as one agent turn.
208 unit + 9/9 F11 host E2E green. (The lone red E2E is a pre-existing,
F11-independent undoMarks timing flake — proven by isolation: it fails
identically with all F11 tests removed.)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Closes the F11 feature: makes the toolbar surface reachable end to end and
guards the edit controls. Per spec §6.4/§6.5/§7.2 SLICE-5.
- package.json: cowriting.showTrackChangesPreview added to editor/title
(when: editorLangId == markdown) — the minimal right-click → Open Review
Preview gateway (#41/#42 expand it later).
- trackChangesPreview: the gateway command accepts the tab's resource Uri
(palette/keybinding still fall back to the active editor); refresh() sends an
`authorable` flag on both render messages; `editControlsEnabled` test seam.
- webview: disable Pin + Ask-Claude on a non-authorable doc (Annotations stays
active — reading is always allowed); RenderMessage.authorable.
- host E2E: the editor/title gateway opens the preview + is markdown-guarded;
edit controls disabled on a non-authorable (read-only-scheme) markdown doc.
- docs: docs/MANUAL-SMOKE-F11.md (live smoke, 10 steps) + README F11 section +
intro line.
205 unit + 53 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A whole-document Ask-Claude rewrite is diffed into hunks and surfaced as N
independent F4 proposals (one per changed hunk) — reusing the F4 single-range
model N times, no new model. Per spec §6.4/§7.2 SLICE-3.
- trackChangesModel: pure `diffToHunks(currentText, rewrittenText)` →
EditHunk[] (vscode-free, deterministic; diffWordsWithSpace, coalescing
adjacent add/remove runs; offsets index currentText).
- trackChangesPreview: `runEditAndPropose(document, target, instruction)` — the
shared host routine (selection → one single-range propose; document → diff →
one propose per hunk; never mutates the doc, INV-10); `askClaude` UI wrapper
(host showInputBox keeps LLM/secrets out of the sealed webview, INV-8/35);
injectable `editTurn` + `setEditTurnForTest` seam (no LLM in CI); the
`askClaude` inbound message branch; `cowriting.editDocument` command for #42
reuse.
- package.json: register cowriting.editDocument, palette-guarded on markdown.
- webview: ✦ Ask Claude to Edit Document button → { askClaude, scope:"document" }.
- unit: diffToHunks fixtures (zero/one/multi-hunk, wholesale, determinism).
- host E2E: stubbed multi-hunk rewrite → N matching proposals, doc untouched;
editDocument command registered + markdown-guarded.
205 unit + 49 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The pure render layer now emits data-src-start/data-src-end (source char offsets
from BlockWithRange) on every LIVE-source rendered block, in BOTH modes — the
contract the webview's selection→source mapping (SLICE-4) walks the DOM for.
Per spec §6.1 INV-36 / §7.2 SLICE-2.
- trackChangesModel: shared `srcAttr(blk)` helper; threaded through renderOp +
renderReviewOp + the renderReview loop (removed blocks → "" / no data-src, as
they have no live source). renderPlain switches from a single whole-document
markdown pass to per-block `<div data-src-start/end>` wrappers — bare divs (no
cw- class) so the off/clean preview stays visually clean while becoming a
selection→source surface. Pure, vscode-free, deterministic (extends INV-22).
- unit (test/trackChangesModel.test.ts): data-src offsets equal
splitBlocksWithRanges in both modes; removed + proposal blocks carry none;
off-mode stays cw--free; determinism.
200 unit + 47 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Homes the orphaned cowriting.pinDiffBaseline command and gives the writer a
reachable Pin control in the preview toolbar. Per spec §7.2 SLICE-1
(docs/superpowers/specs/2026-06-12-f11-preview-toolbar-interaction-surface.md).
- trackChangesPreview: extract onDidReceiveMessage into handleWebviewMessage;
add the F11 `pinBaseline` intent → DiffViewController.pin(previewedDoc) (the
bound doc, not activeTextEditor — §6.7); ToolbarMsg union; receiveMessage test
seam exercising the real message→seam wiring (INV-35).
- webview: ⌖ Pin baseline button in #cw-header posting { type: "pinBaseline" };
theme-aware toolbar-button CSS (light/dark/high-contrast, disabled state).
- package.json: unhide pinDiffBaseline — commandPalette `when` false →
editorLangId == markdown (resolves the #34 orphan from the command side).
- host E2E (test/e2e/suite/f11Toolbar.test.ts): pinBaseline message clears the
change-marks + advances the baseline to `pinned`; palette `when` is reachable.
Also archives the F11 implementation plan to docs/superpowers/plans/.
197 unit + 47 host E2E green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Undo in the editor rendered wrong marks in the F10 review preview: restored
baseline text (and reverted Claude text) was colored as human-authored. Root
cause — attributionController.onDidChange attributed every non-seam document
change to currentAuthor() (always human) and ignored e.reason, so an undo/redo
that re-inserts text created a fresh human span over it. renderReview is pure;
the wrong marks came from these false author spans.
Fix: on e.reason === Undo|Redo, reconcile span geometry but do NOT attribute the
re-inserted chars — an undo is history navigation, not authorship, so restored
text stays neutral (unattributed) rather than falsely claimed by the human.
applyChange gains an `attributeInserted` flag (default true; false on undo/redo);
seam matching is also skipped on undo/redo (the seam only applies forward edits).
Restored text becomes neutral rather than recovering its exact prior provenance
(e.g. undoing a deletion of Claude's text shows it unattributed, not blue) —
perfect restoration would need an attribution history stack synced to the editor
undo stack (fragile due to edit coalescing); filed mentally as a follow-up. The
neutral behavior removes the misleading marks, which is the reported defect.
Tests: E2E reproduces the mid-edit-undo case (buffer stays dirty so the disk-sync
guard doesn't mask it) and asserts restored text is unattributed; +3 unit tests
for the geometric-only applyChange path. 197 unit + 50 E2E green; typecheck clean.
Closes#38
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
F10 (#29) made the rendered preview the single review surface and hid the F6
two-pane vscode.diff view (command + ctrl+alt+d set when:false). This removes
that now-unreachable view code:
- DiffViewController: drop toggle/findDiffTab/epochLabel/isDiffOpen, the
`cowriting-baseline:` TextDocumentContentProvider + BASELINE_SCHEME + baselineUri
+ the content-provider change emitter, and the toggleDiffView command. The
baseline DATA layer is fully intact — ensureBaseline/advance/pin/capture,
getBaseline, baselineFilePath, onDidChangeBaseline, persistence (INV-19), and
the machine-landing auto-advance (INV-18) that F7/F10 consume.
- package.json: remove the toggleDiffView command, its commandPalette entry, and
the ctrl+alt+d keybinding.
- E2E: diffView suite keeps the baseline-data-layer tests, drops the two-pane
view tests; the F10 + no-workspace suites assert toggleDiffView is now absent
(was: declared-but-hidden).
Deliberate deviation from the issue's literal acceptance: pinDiffBaseline is
KEPT. The canonical Solution Design (coauthoring-interactive-review.md §6.7)
scopes the removal to the two-pane VIEW only ("keep the controller + baseline
store"); pin() lives in the baseline lifecycle (§6.4), never touches vscode.diff,
and is exercised by live F7 baseline-reset tests. Where the P3 capture draft and
the approved spec conflict, the spec wins (documentation-leads-automation).
194 unit + 49 E2E green; typecheck + build clean. No F7/F10 behavior change.
Closes#34
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
In the F10 review preview, renderReview now emits each pending proposal whose
anchor resolves as a cw-proposal block immediately after the current-side block
its anchorStart falls in — answering "what is Claude proposing, and where?" by
position — instead of appending all proposals as trailing blocks. Proposals in
the same block are ordered by anchorStart then id (deterministic, INV-33); a
proposal whose anchor doesn't resolve (or precedes all blocks) still trails as a
cw-proposal-unanchored block, never dropped (INV-34).
This implements the inline-at-anchor placement the graduated F10 Solution Design
already specified (coauthoring-interactive-review.md §2/§6.2); the trailing-block
behavior shipped in #29 was a recorded v1 deferral.
Unit: mid-document placement, two proposals in distinct blocks, same-block
ordering determinism, anchored-before-trailing, mixed-set determinism.
E2E: the F10 propose test now asserts the proposal renders before the following
block (in place); accept still lands + clears.
194 unit + 51 E2E green; typecheck clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds test/e2e/suite/f10Review.test.ts covering the F10 flow: open preview
(mode "on", fresh baseline all-unchanged), type → cw-by-human span, propose →
cw-proposal block with ✓/✗, accept → lands + baseline advances + block clears,
reject → vanishes + doc untouched, toggle off → renderPlain has no cw- marks,
status-bar PUC-6 (indicator with no panel, hidden once opened), and the
clean-editor INV-32 facts (retired toggleAttribution, F6 ctrl+alt+d hidden).
Rewrites the obsolete F9 authorship-mode test as an F10 review test
(cw-by-claude in the on-state render; getMode defaults to "on"). Updates
attribution.test.ts (drops the retired isVisible/toggleAttribution toggle,
asserts the toggle is gone) and noWorkspace.test.ts (toggleAttribution +
acceptProposal/rejectProposal are retired, preview-only — INV-32).
Honesty fix in the status-bar seam: hideStatus() clears statusItem.text so
statusText() reports undefined when the indicator is hidden (it previously
returned its stale last value after a panel opened).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Code-quality polish: the webview panel title read 'Track changes: <name>' but
F10 makes the preview the single review surface. Match the 'Open Review Preview'
command rename.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Includes the PUC-6 status-bar indicator (updateStatus + show/dispose hooks)
since the constructor subscription references it (one file, no intermediate
broken build).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Code-quality follow-up: with editor decorations gone (Task 1), the visibility
toggle was dead/misleading — nothing reads `visible`. Remove the field,
toggle()/isVisible(), the cowriting.toggleAttribution registration, and its
package.json command + palette declarations (spec §6.2 retires it). renderActive/
render stay — they refresh the orphan status bar on editor switch.
E2E toggle assertions (attribution.test.ts, noWorkspace.test.ts) are updated in SLICE-4.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per-panel mode (default changes); refresh branches to renderAuthorship reading
AttributionController.spansFor; webview setMode message; extension.ts reorders the
F7 controller after attribution. getMode/setMode test seams.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
PUA sentinel injection through markdown-it; adjacent-span ordering handled;
code/mermaid fences atomic with a block-level author badge.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Spec §6.4: construct GlobalSidecarStore + SidecarRouter (CoauthorStore when root);
remove the no-root early-return + command stubs so F2/F3/F4 are real folder-less;
renderIfOpen gated by isAuthorable; editor-context menus widened to untitled;
export sidecarRouter on CowritingApi. activate now always returns the API.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Spec §6.4: replace the scheme!=file / !isUnderRoot rejections with a single
!isAuthorable branch. isUnderRoot retained as a routing input. Per-condition
messaging (#24) kept for no-editor / empty-selection / non-{file,untitled}.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
BenStullsBets
benstull
Ben Stull