fix: keep research available without search keys
This commit is contained in:
@@ -49,6 +49,8 @@ Tool outputs may be error strings (e.g. search unconfigured, a page that failed
|
||||
Adapt: try another URL or another search angle; if research is genuinely impossible, say so
|
||||
in the brief instead of inventing facts.
|
||||
|
||||
If web_search reports that search is unavailable because credentials are missing, do not retry it with different queries. Use web_fetch when you have source URLs, and continue with any other available sources. If the missing search capability materially limits your work, state the limitation and the missing-key setup needed in your brief. Never invent search results or sources.
|
||||
|
||||
Your FINAL assistant message is your entire deliverable — it must stand alone, using this format:
|
||||
|
||||
## Summary
|
||||
@@ -70,8 +72,7 @@ export interface ResearcherRunOptions {
|
||||
/** The search seam (search port wave, 2026-09-20): resolves the LIVE
|
||||
* provider per call, null = honestly unavailable (the web_search tool
|
||||
* reports that in its output — no crash, no silent fallback). Absent
|
||||
* seam reads as unavailable: the server wiring always injects one; the
|
||||
* dispatch-level drop (jobTools) keeps doomed jobs from even starting. */
|
||||
* seam reads as unavailable; web_fetch remains usable without it. */
|
||||
search?: () => SearchProvider | null;
|
||||
/** The researcher's PDF dir (PDF wave, 2026-10-07): when present, the
|
||||
* toolset gains web_fetch_pdf + pdf_info/pdf_extract/pdf_search scoped
|
||||
|
||||
Reference in New Issue
Block a user