zan-browser 3.0.47 → 3.0.48
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/dist/agent/prompt.js +1 -1
- package/package.json +1 -1
package/dist/agent/prompt.js
CHANGED
|
@@ -38,7 +38,7 @@ At every step, reason about what you have and what gets you closest to the data:
|
|
|
38
38
|
- If network traffic looks complex, encoded, or uses GraphQL/persisted queries — use analyze_network_logs with decode_base64=true to decode base64 params, extract operationNames, and reveal the real API structure. Then call the platform's public REST API directly if one exists.
|
|
39
39
|
- If a site has no useful XHR traffic, look at its JS bundle URLs (visible in script tags via eval_js). Use extract_endpoints on those URLs to find internal API patterns.
|
|
40
40
|
- Use fetch_url only when you have a concrete URL from evidence — network logs, JS bundles, or documentation. Never guess.
|
|
41
|
-
- If fetch_url returns 403 on an endpoint
|
|
41
|
+
- If fetch_url returns 403 on an endpoint visible in network logs with relevant data, use get_cookies to get session cookies, then retry fetch_url with header "Cookie: <cookieHeader>". Only use get_cookies as a fallback after a 403, not proactively.
|
|
42
42
|
- If data is server-rendered with no API, use scrape or extract_dom as a last resort.
|
|
43
43
|
- If the current URL contains the seed value in the path and the page has visible listing data in the DOM, that URL is a valid parameterizable endpoint. Use scrape to extract the data and set data_found to true — do not keep searching for a JSON API that may not exist.
|
|
44
44
|
|