You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
New cookbook: browsers/telemetry/pausing-for-captcha-solves.mdx — pauses a browser-loop agent when captcha_solve_started fires and resumes once captcha_solve_result / captcha_challenge_result reports a terminal outcome, reusing the abort-and-resume pattern from the existing Playwright/computer-use fallback cookbook.
Wired into docs.json under Telemetry, and cross-linked from the fallback cookbook's Next Steps.
Documents the @onkernel/sdk version conflict between browser-loop's pinned dependency (predates the telemetry API) and the SDK version the script needs, with a verified overrides fix.
Test plan
docs.json and the doc's JSON snippet parse as valid JSON
The full TypeScript example type-checks against the real @onkernel/browser-loop, @onkernel/sdk, and @earendil-works/pi-agent-core packages (installed at the exact versions browser-loop depends on, with the overrides fix applied) — caught and fixed a real API mismatch (InMemorySessionRepo/AgentHarnessEvent from a newer pi-agent-core than browser-loop pins) and a missing-telemetry-on-old-SDK issue before they shipped
Not run end-to-end against a live Kernel browser (no API key in this environment) — the script is a straightforward extension of the already-published fallback cookbook's pattern, but hasn't been observed solving a real captcha
Note
Low Risk
Documentation and nav/cross-links only; no runtime or API code changes in the repo.
Overview
Adds a Telemetry cookbook that shows how to pause a @onkernel/browser-loop agent while Kernel’s stealth captcha solver runs, using captcha telemetry plus a live DOM probe instead of screenshot-only prompt instructions.
The doc walks through a full captcha-gate.ts example: stream captcha_* events with bounded waits and task_id/challenge_id correlation, resolve outcomes separately for solver tasks vs visible challenges, and hook AgentHarnesstool_call to hold or block actions with explicit verdict messages. It also covers setup (pinned pi-agent-core, ESM), example agent messaging, and known limits (press-and-hold probe gap, unordered events).
Navigation: registers the page in docs.json under Telemetry and links it from the Playwright computer-use fallback cookbook’s Next steps as an alternative mid-run hold pattern.
Reviewed by Cursor Bugbot for commit 6283404. Bugbot is set up for automated code reviews on this repo. Configure here.
Shows how to use browser-loop's abort/resume pattern with the
captcha_solve_started / captcha_solve_result / captcha_challenge_result
telemetry events, so an agent stops acting while Kernel's solver is
working and resumes once it reports a terminal outcome.
Buffers a terminal telemetry event that arrives before waitForOutcome
is called, clears tracked IDs on timeout so the next captcha still
triggers a pause, stops a same-challenge task from overwriting the
tracked task_id, and distinguishes a task-level success from a
challenge actually clearing.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 3712d40 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +252 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after approval:3712d40 only edits the cookbook example — outcome buffering, distinct challenge_solved vs task_success, same-episode ID handling, and timeout ID reset. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit narrows example-code correctness; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
If an episode's outcome buffers before waitForOutcome is called and a
new episode starts before the driver gets back to it, the buffer is
now cleared so the driver waits on the current episode instead of
resuming on the stale one. Also breaks the dense gate-behavior
paragraph into bullets for scannability.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 9df87e2 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +261 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:9df87e2 only edits the cookbook — clears a stale buffered outcome when a new captcha episode opens, and splits the gate-behavior paragraph into bullets. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit tightens example-code correctness; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
Delivery is unordered, so a captcha_solve_started for an episode can
arrive after that episode's own terminal result already buffered.
The gate now remembers which task_id/challenge_id a buffered outcome
belongs to, so a late duplicate for that same episode is recognized
and left alone instead of being mistaken for a new episode (which
previously discarded the buffer and re-triggered a spurious pause).
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD d7f6e03 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +286 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:d7f6e03 only edits the cookbook example — keeps the buffered outcome's task_id / challenge_id so a late duplicate captcha_solve_started is ignored instead of clearing the buffer. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit tightens example-code correctness; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
Starts on duckduckgo.com, navigates to 2captcha.com's public
Cloudflare Turnstile demo, and has the agent summarize the page's
explanation of how Turnstile is solved once the challenge clears.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 0ed73cc against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +290 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:0ed73cc only edits cookbook demo text — switches the example target from the public Google reCAPTCHA v2 demo to duckduckgo.com → 2captcha.com's Cloudflare Turnstile demo, and updates the example TASK_PROMPT / comment to match. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example TASK_PROMPT / systemPrompt text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit is demo-URL and example-prompt wording only; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
Removes inline references to the Playwright/computer-use fallback
cookbook from the body text (kept only in Next Steps), and replaces
the concrete duckduckgo/2captcha demo task with a placeholder the
reader fills in with their own task.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 911955e against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +287 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:911955e only edits cookbook copy — drops inline links to the Playwright/computer-use fallback page from the body (kept in Next Steps) and replaces the concrete duckduckgo/2captcha demo TASK_PROMPT with a placeholder. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example TASK_PROMPT / systemPrompt text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit is wording and a placeholder prompt; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 924f68a against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +293 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:924f68a only adds a markdown table in the cookbook intro listing the three captcha event types and their outcomes. Same files, still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example TASK_PROMPT / systemPrompt text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit is a reference table; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
browser-loop's exact @onkernel/sdk pin is relaxed to a caret range
(kernel/browser-loop#94), matching the same removal already done for
the Playwright/computer-use fallback cookbook (#548).
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 738ec65 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +279 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:738ec65 only deletes a <Warning> callout about a stale @onkernel/sdk version-pin / overrides workaround. Same files, still documentation and nav config. Surface area shrank slightly.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example TASK_PROMPT / systemPrompt text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental commit is copy removal only; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
The previous gate paused on captcha_solve_started, joined challenge
results only when a start had carried a challenge_id, and had no
fallback when an event never arrived. On a real reCAPTCHA that dropped
the challenge-level outcome entirely and re-prompted the agent once per
solver task.
The gate now pairs tasks on task_id only, records every challenge result
and marks whether it can be attributed, bounds every wait, and falls
back to a read-only page probe. It holds at the tool_call hook instead
of aborting the turn, so an action is stopped before it reaches the
page and no re-prompt is needed.
Split into four named steps with the complete file in an accordion, and
pin pi-agent-core to the version browser-loop depends on -- the previous
install line resolved a newer one that the script doesn't compile
against.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 4d7645f against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +522 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:4d7645f rewrites the cookbook around captcha correlation rules — task-id joins, bounded waits, a read-only page probe, and a tool_call hold instead of abort-and-resume. Same three files; still documentation and nav config. No new paths, CI, or runtime code.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental rewrite enlarges the example but does not expand blast radius beyond this docs site.
Approval from the prior Very Low assessment remains in effect.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD dbdabaa against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +518 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:dbdabaa only folds a <Note> callout into body prose in the cookbook intro. Same three files; still documentation and nav config. No new paths, CI, or runtime code.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental edit is copy presentation only; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD c694177 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +518 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:c694177 is a merge of main into this branch. The captcha cookbook, nav entry, and cross-link are unchanged vs dbdabaa. Incoming main edits (info/pricing.mdx, apps/invoke.mdx) are already on main and are not new surface from this PR. Same three files vs main; still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Merge from main does not expand this PR's blast radius.
Approval from the prior Very Low assessment remains in effect.
The reason will be displayed to describe this comment to others. Learn more.
reviewed and QA'd the cookbook end to end. the overall structure and telemetry/task distinction are clear, but the copied example has a few correctness and setup issues:
bugs
browsers/telemetry/pausing-for-captcha-solves.mdx:34-41 — the exact clean-directory instructions fail before main() with ERR_PACKAGE_PATH_NOT_EXPORTED: tsx treats the .ts file as CommonJS while pi-agent-core exposes ESM. add npm pkg set type=module or use captcha-gate.mts.
browsers/telemetry/pausing-for-captcha-solves.mdx:98-125,272-275 — the detached telemetry stream rejects when normal cleanup deletes the browser; QA produced an unhandled 410 browser session no longer exists and exit code 1. cancel/close the stream before deletion or catch the expected session-gone error.
browsers/telemetry/pausing-for-captcha-solves.mdx:165-170 — pending() excludes successful task-only results and unresolved tasks after their deadline. in either case the next action bypasses resolve() and the page probe, breaking the documented Turnstile and missing-result paths. treat every unconsumed task record as pending until reset.
browsers/telemetry/pausing-for-captcha-solves.mdx:210-224,260-267 — tokenPresent is collected but ignored. solved reCAPTCHA widgets commonly remain visible, so the gate can block an action after a response token exists. use the token in the blocking decision and include [name="h-captcha-response"].
reliability and docs
browsers/telemetry/pausing-for-captcha-solves.mdx:250-267 — all telemetry events can be absent, but the generic system prompt supplies no fallback and the page is probed only after telemetry populates the gate. retain the automatic-captcha wait instruction or probe before every relevant action.
browsers/telemetry/pausing-for-captcha-solves.mdx:30-35 — the command saves "^0.83.0", not an exact pin. use --save-exact if newer versions are known to break these imports.
browsers/telemetry/pausing-for-captcha-solves.mdx:210-224,503-508 — press-and-hold has challenge telemetry but no page selector. an unresolved challenge is logged and allowed through rather than falling back to a telemetry verdict for the agent. add detection or clarify the limitation.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 73f8d6e against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +518 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment:73f8d6e is a merge of main into this branch. The captcha cookbook and Next-steps cross-link are unchanged vs c694177. Incoming main edits (Cookbooks tab, changelog, analytics pixel, network-access page, vault/payments copy) are already on main and are not new surface from this PR. Same three files vs main; still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Merge from main does not expand this PR's blast radius.
Approval from the prior Very Low assessment remains in effect.
@earendil-works/pi-agent-core ships ESM only, so tsx resolves the
script as CommonJS and fails immediately without "type": "module" in
package.json. Reproduced the exact error following the documented
setup steps as written, then verified npm pkg set type=module fixes
it.
The gate's telemetry loop runs detached and was never explicitly
stopped, so it keeps its connection open past the browser's deletion.
Reproduced: the documented script hangs indefinitely after finishing
its task instead of exiting (confirmed via a live run, deleteByID
completes fine but the process never returns). Fixed by saving the
stream's AbortController and calling gate.close() in main()'s finally
block before deletion; verified the same live run now exits cleanly
in ~5s.
pending() only counted a settled task if its status was a non-success
terminal value, so a successful task-only result (Turnstile's usual
path, per the cookbook's own documented example) or a task whose
deadline lapsed with no result at all were neither holding() nor
pending() -- resolve() and the page probe never ran for them, and the
agent was never told. Replaced the status-based check with the same
isOpen() predicate openTasks() already uses, so pending() means "some
task settled, one way or another." Verified against a scripted
turnstile-success case: it now reproduces the exact message documented
in the "What the agent is told" table, which the prior logic could
never actually produce.
A solved reCAPTCHA or hCaptcha widget commonly stays rendered on the
page, so gating solely on verdict.page.widgets.length blocks the
agent indefinitely after the challenge already produced a usable
response token. The block decision now also clears when a token is
present. Also added hCaptcha's response field name -- the probe was
only checking Turnstile's and reCAPTCHA's -- so tokenPresent can
actually see an hCaptcha solve.
The gate only interrupts the agent when telemetry gives it something
to act on, but every captcha event can be absent. Without a fallback,
a captcha telemetry misses entirely leaves the model with no guidance
at all. Added the same automatic-solve wait instruction already
documented for the no-telemetry case elsewhere, so there's a floor
under the gate rather than nothing. Verified the addition doesn't
change behavior on a normal, captcha-free run.
The Warning says to pin @earendil-works/pi-agent-core, but
"pkg@0.83.0" in a plain npm install still saves a caret range
(^0.83.0), which a patch release could still break against. Split it
into its own npm install --save-exact call, verified it now writes a
literal "0.83.0" with no operator while @onkernel/sdk stays on a
flexible range, and confirmed the resulting install still runs the
script end to end.
Press-and-hold emits captcha_challenge_result like reCAPTCHA and
hCaptcha, but PROBE's selector groups have no entry for it, so
widgets always comes back empty for it and the tool-call handler lets
the agent proceed even while the challenge is genuinely still open.
Documented the gap explicitly rather than leaving it implied by the
generic "best-effort" bullet, since it's a real class of false
negative, not just an edge case of missed markup.
Thanks for the thorough QA — fixed all 7, each as its own commit:
Bugs
ERR_PACKAGE_PATH_NOT_EXPORTED (454748b) — reproduced exactly as described (pi-agent-core is ESM-only, tsx was resolving it as CJS). Setup now runs npm pkg set type=module; verified from a clean install.
Detached telemetry stream on cleanup (b6ef142) — couldn't reproduce your exact 410/exit-1, but confirmed the underlying defect directly: the documented script hangs indefinitely after finishing its task instead of exiting (deleteByID completes fine, the process just never returns). Fixed by saving the stream's AbortController and calling gate.close() before deletion in main()'s finally; verified a live run now exits cleanly in ~5s instead of hanging.
pending() dropping results (5f8899b) — reproduced with a scripted turnstile-success case: pending() stayed false for it, so resolve() never ran and the documented "task-only" message could never actually be produced. Replaced the status-based check with the same isOpen() predicate openTasks() uses (covers success and expired-without-result too); re-ran the scripted case and it now reproduces the exact message from the "What the agent is told" table.
tokenPresent ignored (669abf9) — the block decision now also clears on tokenPresent, and added hCaptcha's h-captcha-response to the selector (was only checking Turnstile's and reCAPTCHA's).
Reliability/docs
System-prompt fallback (c1066d8) — added the same automatic-solve wait instruction as a floor for when telemetry misses everything.
--save-exact (a2af5b3) — split into its own npm install --save-exact call for just pi-agent-core, verified it now writes a literal 0.83.0 while @onkernel/sdk stays on a flexible range.
Press-and-hold selector gap (778f23b) — went with clarifying the limitation rather than guessing at a selector I couldn't verify; documented explicitly that it emits challenge telemetry but is never actually held, since PROBE has no marker for it.
Verified end to end with a live Kernel browser + real agent run after each code fix (typecheck + live run, not just reading it), and once more with all seven combined.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 778f23b against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +547 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment (73f8d6e): Only the cookbook MDX (+43 / -14). Setup now uses --save-exact for pi-agent-core plus type=module; the example closes the telemetry stream before browser delete, treats every settled task as pending(), uses tokenPresent in the block decision (including hCaptcha), adds an example systemPrompt fallback, and documents that press-and-hold has no probe selector. Same three files vs main; still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental edits tighten example-code correctness and copy; they do not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
reset() unconditionally cleared every task, so a follow-up solver
task that started during resolve()'s challenge-grace wait (only
watching for captcha_challenge_result there, not new tasks) got wiped
out while still genuinely in flight -- holding() went false right
after, letting the agent act while Kernel's solver was still working.
Reproduced with a scripted case (a task still open, no covering
challenge result, at the moment reset() runs) showing holding()
incorrectly flip to false; fixed by only clearing a task once
something actually covers it -- its own terminal status, or a
challenge result for its episode. Verified the fix against that case
and a regression case where a challenge result legitimately does
cover a leftover open task (multi-task-same-challenge), which still
clears correctly.
reset() cleared every task_id and challenge_id it consumed, so a
captcha_solve_started that arrives after its own result -- delivery
is unordered, this is explicitly possible -- gets treated as a
brand-new task in the now-empty map, and the next tool call holds for
a full TASK_SETTLE_MS waiting on a result that was already seen and
reported. Added a settled set of IDs reset() has cleared; a late
event matching one is dropped instead of reopening it. Reproduced the
phantom-reopen with a scripted late-duplicate-start case and verified
it's now ignored, plus regression-checked the covered-leftover-task
and task-only-success cases.
The reason will be displayed to describe this comment to others. Learn more.
Stale comment
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD e394a41 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +579 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment (778f23b): Only the cookbook MDX (+43 / -11). Example reset() now keeps an uncovered in-flight task, and a settled set drops late events for IDs already reported. Same three files vs main; still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental edits tighten example-code correctness; they do not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
reset() permanently blacklisted every reported task_id and
challenge_id, but Kernel can reuse a challenge_id when an episode
continues across a page reload -- a genuinely new outcome for one
would be silently dropped forever, and a new task under it would
still cost a full CHALLENGE_GRACE_MS wait for a result that could
never be recorded. Replaced the permanent set with a map of when each
ID was settled; a duplicate is only dropped within TASK_SETTLE_MS of
that, then treated as fresh again. Verified the original short-window
duplicate is still blocked, a reload's new outcome past the window is
no longer dropped, and reran the existing regression scenarios
(task-only success, follow-up task survives, covered leftover task).
Documented the resulting edge case (a reload within TASK_SETTLE_MS of
the prior verdict can still be missed) in Limits.
The reason will be displayed to describe this comment to others. Learn more.
Risk assessment: Very Low
Verdict: Risk did not increase. Prior approval stands (not re-approving).
Re-evaluated HEAD 6283404 against the previously approved commit 8e9d21e from the file diffs only (not the PR description).
Current scope (3 files, +596 / -1 vs main):
File
Change
browsers/telemetry/pausing-for-captcha-solves.mdx
New Mintlify cookbook (example TypeScript + setup notes)
docs.json
Adds that page under the existing Telemetry nav group
browsers/playwright-computer-use-fallback.mdx
One Next-steps cross-link
What changed after the last assessment (e394a41): Only the cookbook MDX (+36 / -19). Example settled Set is now a settledAt Map with a TASK_SETTLE_MS TTL, plus a Limits note that a reload inside that window can still drop a reused challenge_id. Same three files vs main; still documentation and nav config.
Why still Very Low
Documentation and Mintlify nav only. No application, API, auth, billing, or infrastructure code.
No CODEOWNERS file in this repo, so no required owner review.
Example systemPrompt / cookbook text lives in user-facing docs, not in model-instruction or production prompt files.
Incremental edit tightens example-code correctness; it does not expand blast radius.
Approval from the prior Very Low assessment remains in effect.
Sent by Cursor Automation: Assign PR reviewers
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
rgarcia
left a comment
Summary
browsers/telemetry/pausing-for-captcha-solves.mdx— pauses abrowser-loopagent whencaptcha_solve_startedfires and resumes oncecaptcha_solve_result/captcha_challenge_resultreports a terminal outcome, reusing the abort-and-resume pattern from the existing Playwright/computer-use fallback cookbook.docs.jsonunder Telemetry, and cross-linked from the fallback cookbook's Next Steps.@onkernel/sdkversion conflict betweenbrowser-loop's pinned dependency (predates the telemetry API) and the SDK version the script needs, with a verifiedoverridesfix.Test plan
docs.jsonand the doc's JSON snippet parse as valid JSON@onkernel/browser-loop,@onkernel/sdk, and@earendil-works/pi-agent-corepackages (installed at the exact versionsbrowser-loopdepends on, with theoverridesfix applied) — caught and fixed a real API mismatch (InMemorySessionRepo/AgentHarnessEventfrom a newerpi-agent-corethanbrowser-looppins) and a missing-telemetry-on-old-SDK issue before they shippedNote
Low Risk
Documentation and nav/cross-links only; no runtime or API code changes in the repo.
Overview
Adds a Telemetry cookbook that shows how to pause a
@onkernel/browser-loopagent while Kernel’s stealth captcha solver runs, using captcha telemetry plus a live DOM probe instead of screenshot-only prompt instructions.The doc walks through a full
captcha-gate.tsexample: streamcaptcha_*events with bounded waits andtask_id/challenge_idcorrelation, resolve outcomes separately for solver tasks vs visible challenges, and hookAgentHarnesstool_callto hold or block actions with explicit verdict messages. It also covers setup (pinnedpi-agent-core, ESM), example agent messaging, and known limits (press-and-hold probe gap, unordered events).Navigation: registers the page in
docs.jsonunder Telemetry and links it from the Playwright computer-use fallback cookbook’s Next steps as an alternative mid-run hold pattern.Reviewed by Cursor Bugbot for commit 6283404. Bugbot is set up for automated code reviews on this repo. Configure here.