The Document PiP window had no close/return entry once popped out —
users could not dismiss it from inside the window. Add an onClosePip
callback and render a close button in the PiP header; clicking it
calls pipWindow.close() and the content returns to its carrier.
The observation thumbnail publisher hardcoded mediaType as "image/png"
regardless of the actual capture bytes. Reuse the existing
sniffImageMediaType helper (exported from image.ts) so the HTTP
thumbnail endpoint declares the correct Content-Type when Chromium
returns JPEG captures.
Closes#153 (Bug 2 fully; Bug 1 was fixed in #135, this covers the
remaining observation-thumbnail path).
Co-authored-by: Cursor <cursoragent@cursor.com>
browser_screenshot declared every capture as image/png regardless of the
actual bytes. On Chromium/Edge builds where captureVisibleTab returns JPEG
bytes, the stored attachment reference recorded mediaType "image/png" against
JPEG bytes, so dsh-attachment-local's read-time metadata verification fails
with ATTACHMENT_CORRUPT ("Stored attachment metadata does not match its
reference"), surfacing as code UNKNOWN and triggering endless LLM retries.
- image.ts: sniff the capture's magic bytes (PNG/JPEG/WebP/GIF) and declare
that real media type to attachments.saveImage, instead of always "image/png".
- tools.ts: assemble the tool result from ref.mediaType (the store's verified
type) rather than re-declaring "image/png" as const, and relax the output
schema's const to the accepted media-type enum.
- tests: add a regression case asserting JPEG capture bytes surface as
image/jpeg.
Co-Authored-By: DeepSeek Harness <noreply@deepseek.com>
Add a new monorepo package that registers BrowserSkill (bsk) browser
automation as model-visible tools in DeepSeek Harness (dsh):
- 11 tools mapping to bsk CLI --json commands: session start/stop/list,
navigate, snapshot, observe, click, fill, press, screenshot, emulate
- multi-session support: optional session arg with a current-session
pointer, session id echoed in every result, configurable concurrency
cap (default 5), and dispose-time cleanup of plugin-started sessions
- cancellation: exec.signal aborts kill the underlying bsk child process
- terminal-style UI cards (command line as the call card, output as the
result card); screenshots inline the PNG through the host attachment
store when the model route accepts image input, else return the path
- config: bskPath / defaultTimeoutMs / maxSessions via Schemastery schema
- install guidance when the bsk binary is missing; bsk JSON error
envelopes (code/message/hint) surfaced to the model
- unit tests mock the bsk runner (no real browser required); the root
lint script now also runs the package's typecheck + tests so the
existing CI frontend job covers the new package (a dedicated CI job
diff is provided in the PR; the pushing token lacks workflow scope)
