Every tool.observe moved the real cursor across the page to look for
hover-only menus and tooltips. Measured against real Chrome, that cost a
median of 2.5s and up to 4.7s per observation — 73% to 99% of total
observe wall clock — and returned 3 hits across 12 pages. A positive
control confirms the probes work; the yield on real pages simply does not
justify putting them on the default path.
observe now probes only when asked, via `probe_hover` (`--probe-hover`).
Agents that know which element hides a menu should keep using
`bsk hover <ref>`, which is both cheaper and precise. A held hover latch
still suppresses probing, since probing would move the cursor off the
element the caller is deliberately holding.
Correctness fixes that apply whether or not probing is requested:
- Hover probing ran between the DOM snapshot and the accessibility tree
fetch, so a hover that opened a menu left one half of an observation
describing the page before the change and the other half after it. Both
chains now run after all capture is complete.
- The overlay bypass toggled once per chain and swallowed restore
failures, which could leave the agent overlay hidden for the rest of
the tab's life. It is now reference counted across both chains, and a
failed restore is reported instead of discarded.
- Observations report whether they hovered the page and whether that
revealed content the static tree lacks, so callers can judge how far
the live page may have drifted from the returned snapshot.
Probe budgets were checked at the top of the loop, so a candidate could
start with 1ms remaining and still run a full settle window, overshooting
by up to 634ms. Candidates are now admitted only if they fit. Raising
MAX_HOVER_PROBE_MS to 2600 is part of that fix rather than a relaxation:
2000 was never the real ceiling, and enforcing it literally would have
silently dropped a candidate the positive control depends on.
Tooltip misses are cached too, with a shorter TTL than hits, so repeat
observations of the same page no longer re-probe every unlabelled control
from scratch.
Add `bsk network` as a sibling of the `console` command from #7,
reading the buffered network responses / failures for a tab. Fills the
remaining half of the request in #2 (console landed in #7; network did
not).
Mirrors the console command's conventions exactly:
bsk network --session <id> [--tab-id N] [--since C]
[--limit N] [--max-text-chars N]
The extension enables `Network` on attach (best-effort) and buffers
`responseReceived` / `loadingFailed` events per tab with a monotonic
sequence, correlating them to `requestWillBeSent` for method/URL. Reads
are cursor-paginated (`since` -> `next_since`) and bounded (`limit`,
`max_text_chars`) for agent-context safety, same as console. Entries
carry the real HTTP status code (response) or the CDP failure reason
(failure) — things `performance.getEntriesByType('resource')` can't
give.
Classified read-only (pass-through under the pending-interrupt gate);
defaults to the Agent Window's active tab.
Wires bsk-protocol (Method + params/result/entry structs + schema),
bsk-cli (network subcommand + daemon dispatch allow-list) and the
extension (handler + dispatcher route), with unit / parse / IPC tests
paralleling the console command.
Refs #2.
drakezhang