| name | chatgpt-web |
| description | Query logged-in ChatGPT through one BrowserOS ChatGPT tab after applying the canonical $browseros operating contract and shaping the prompt with $prompt-authoring discipline. Use when the user explicitly wants the ChatGPT web provider/capability, optional attachments, or an exact existing ChatGPT conversation continued. New clean conversation is the default; exact-conversation continuation must be explicit. Defaults to Pro with Extended thinking, runs serially, and waits patiently. Not for OpenAI API work, generic browser automation, automated login, scripts, runners, or hidden harnesses. |
| metadata | {"short-description":"Query logged-in ChatGPT through BrowserOS"} |
ChatGPT Web
Use this skill when the user wants a head start querying ChatGPT through the
already logged-in BrowserOS browser session.
This is a prose-only helper skill. It uses BrowserOS MCP directly. It ships no
scripts, runners, controllers, harnesses, schemas, or automation infrastructure.
Read and apply ../browseros/SKILL.md before the first BrowserOS call. The
canonical BrowserOS skill owns page provenance, lifecycle, proof, recovery,
secrets, and cleanup; this skill owns the ChatGPT-specific workflow.
Read ../_shared/agent-orchestration-policy.md before the query. ChatGPT Web is
an intentional provider/browser-capability lane rather than a generic local
child-agent route. Starting a new conversation and continuing an exact existing
conversation are different context choices; never inherit whatever conversation
happens to be open without deciding which one the user wants.
Use When
- The user asks to ask ChatGPT, consult ChatGPT, get ChatGPT's opinion, or run a
prompt through the ChatGPT web UI.
- The user wants BrowserOS MCP to drive logged-in ChatGPT instead of using the
OpenAI API.
- The user wants local attachments included in a ChatGPT web prompt.
- The user has a rough prompt and wants it shaped before sending.
Do Not Use When
- The user wants OpenAI API usage or product/API guidance.
- The user wants generic web automation unrelated to ChatGPT.
- BrowserOS MCP is unavailable in the current host.
- ChatGPT is not already logged in in BrowserOS.
- The task would require automated login.
- The user requests more than 10 attachments.
Non-Negotiables
- Apply
$browseros before the first direct BrowserOS MCP call and obey it
throughout this workflow.
- Use BrowserOS MCP, not
web.run, OpenAI API calls, shell browser scripts, or
direct cookie/session handling.
- Use one BrowserOS
https://chatgpt.com/ tab for the whole run. Reuse an
eligible current-agent-controlled ChatGPT page when one can be safely
task-adopted under $browseros; otherwise open exactly one ChatGPT page.
Reusing the page does not mean reusing its conversation. Do login check,
conversation selection, mode selection, attachment upload, submission,
waiting, and response reading in that same page.
- Do not open extra ChatGPT tabs for polling, attachment handling, retries,
separate prompts, or readback.
- Run ChatGPT Web prompts serially. If the user gives multiple ChatGPT asks,
process them one at a time in the same ChatGPT tab. Keep them in one
conversation only when they are explicit follow-ups; otherwise start a new
clean conversation in that tab for each independent ask. If the user asks for
parallel ChatGPT Web runs, explain that this skill runs serially to avoid web
session rate limits, then proceed sequentially. If simultaneous ChatGPT Web
runs are mandatory, fail loudly instead of opening parallel tabs.
- Apply
$prompt-authoring to the actual populated prompt before every
submission. Preserve the user's intent and any explicitly requested verbatim
relay; otherwise remove hidden caller assumptions, leading success criteria,
and closed evidence paths before sending.
- Default to ChatGPT
Pro with Extended thinking when the user does not name
a mode or effort.
- Respect explicit user choices for
Instant, Thinking, Pro, Light,
Standard, Extended, or Heavy.
- Do not downgrade or upgrade the requested mode silently.
- Never automate login. If ChatGPT is not logged in, fail loudly and tell the
user to log in manually in BrowserOS.
- Treat
Pro, especially Extended or Heavy, as a long-running mode. A
response can take 10+ minutes. Wait patiently in the same tab until generation
finishes before reading or submitting anything else.
- Do not print, save, summarize, or inspect account details, cookies, tokens,
raw session payloads, or other secrets.
- Enforce a maximum of 10 attachments. Do not silently drop files.
- Keep the result simple: ChatGPT's answer plus a short receipt.
- Default to a new clean ChatGPT conversation. Continue an exact conversation
only when the user asks to continue it and the intended conversation can be
identified. If it cannot, stop and ask for the missing conversation choice
rather than sending into an unrelated history.
First Move
- Resolve the user's desired ChatGPT ask and any explicit mode, effort, model,
or attachment requests.
- Apply
$prompt-authoring to the populated prompt before touching ChatGPT.
Keep it faithful to the user's intent, preserve an explicitly requested
verbatim relay, and make caller hypotheses challengeable rather than task
truth.
- Resolve
conversation = new-clean | continue-exact. Default to new-clean;
use continue-exact only from an explicit user request and an identifiable
target conversation.
- Under
$browseros, select the single current-agent-controlled ChatGPT page
for the run: safely task-adopt an eligible https://chatgpt.com/ page, or
open exactly one new page.
- Verify that page is logged in before doing anything else.
- In that page, open a new chat for
new-clean, or navigate to and verify the
exact requested conversation for continue-exact. Do not submit while the
page is merely showing an arbitrary prior thread.
Login Check
From the ChatGPT page, use BrowserOS MCP page JavaScript to fetch:
fetch('/api/auth/session', { credentials: 'include' })
Use only the safe boolean result: whether the parsed JSON has a user value.
Do not display or store the returned user, account, token, cookie, or session
fields.
If the session does not prove a logged-in user, stop with:
BrowserOS is not logged in to ChatGPT. Open https://chatgpt.com/ in BrowserOS,
log in manually, then rerun $chatgpt-web.
If the endpoint cannot be checked, fail loudly. Do not infer login from visible
page controls.
Mode And Effort
Default when the user does not specify:
mode = Pro
effort = Extended
model family = current ChatGPT default
Use the ChatGPT model pill beside the composer. Prefer Configure... when
available because it exposes the Intelligence dialog with explicit model
options and thinking effort.
Observed controls to select from:
- mode:
Instant, Thinking, Pro
- effort:
Light, Standard, Extended, Heavy
- model family: leave as current unless the user explicitly names one
Do not run a Pro prompt merely to test the skill. Only use Pro when the user's
actual request needs the default or explicitly asks for it.
Because the default is Pro with Extended thinking, the default path also
requires patient waiting. Do not treat a long silent period as failure by itself.
Attachments
Preflight attachments before browser interaction:
- every path must be absolute
- every path must exist
- count must be 10 or fewer
Use the BrowserOS file-upload path that works with ChatGPT in the selected
ChatGPT page:
- Create a temporary visible file input in the page for BrowserOS MCP to use.
- Snapshot the temporary input and use the live BrowserOS MCP
upload tool
with its exact ref and the user's absolute paths.
- Transfer the selected
File objects into ChatGPT's hidden #upload-files
input.
- Dispatch
input and change events on the ChatGPT input.
- Confirm visible attachment chips by filename.
- Remove the temporary visible input.
If any requested filename does not appear as an attachment chip, stop before
submitting.
Submission
- Verify the selected conversation mode one final time. For a new clean run,
the page must be a new chat; for continuation, the visible thread must be the
exact requested conversation.
- Fill the ChatGPT composer with the final prompt.
- Confirm the selected mode and effort match the request or default.
- Confirm every attachment chip is present.
- Click
Send prompt.
- Wait in the same tab until generation finishes. For
Pro, Extended, or
Heavy, 10+ minutes can be normal; poll slowly and let ChatGPT finish.
- Do not refresh, resubmit, open another tab, or start another ChatGPT prompt
while a response is still generating.
- Treat failure as concrete, not time-based: visible ChatGPT error, lost
login/session, required manual user action, missing attachment before send,
or a clearly inactive page with no generation indicator and no response
progress after a patient wait.
- Read the latest assistant response from the page.
Output
Return:
- ChatGPT's answer
- mode and effort used
- conversation mode used (
new-clean or continue-exact)
- attachment filenames, if any
- a short note if the prompt was shaped before submission
- a short note when the run waited for a long Pro response
If the run fails, name the exact failed condition and the next manual repair.
