ai-browser-bridge 0.4.0 → 0.5.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -149,6 +149,36 @@ Paths that escape the repo root are skipped; files over 100 KB are summarized ra
149
149
  selects Chats, and returns matching Conversations across titles and message content. It scrolls
150
150
  the result overlay up to `--limit`; add `--open` to navigate the browser to the best match.
151
151
 
152
+ **Rate-safe Project organization** — prepare one ordered plan with Conversations for any number
153
+ of Projects (Conversation ids are preferred because they can be verified exactly):
154
+
155
+ ```json
156
+ [
157
+ { "conversation": "11111111-1111-4111-8111-111111111111", "project": "Yoga App" },
158
+ { "conversation": "22222222-2222-4222-8222-222222222222", "project": "Invoices & Purchases" }
159
+ ]
160
+ ```
161
+
162
+ Then validate it and start the autonomous queue:
163
+
164
+ ```bash
165
+ bridge chat organize --plan @organization-plan.json --dry-run
166
+ bridge chat organize --plan @organization-plan.json \
167
+ --interval 60-90 --cooldown 600 --max-attempts 4
168
+ ```
169
+
170
+ The queue verifies every move from ChatGPT's loaded Project label, with the destination Project page
171
+ as a fallback. It spaces UI operations by a fixed interval or randomized range; the defaults are a
172
+ 60-second interval, a 600-second rate-limit cooldown, and four attempts per Conversation. For large
173
+ history cleanups, the randomized `--interval 60-90` profile above is recommended. The queue
174
+ acknowledges ChatGPT's “Too many requests” modal before cooling down, and automatically relaunches a
175
+ closed browser session. Neither infrastructure condition consumes a move attempt. Progress is
176
+ persisted under `.bridge/chat-organization-queues/`; rerunning the same plan resumes pending work
177
+ without repeating completed moves. Once the queue has started and processed or deferred one item,
178
+ it can continue unattended without an agent watching it. Use `--restart` only when you intentionally
179
+ want to replay the full plan. `--json` keeps the final summary on stdout and writes progress to
180
+ stderr.
181
+
152
182
  ## Agents & fan-out
153
183
 
154
184
  `bridge ask` is a stable, non-interactive surface for scripts and agents — it prints to stdout, never prompts in a pipe, and `--json` emits a machine-readable payload.