dguiducci/skald-connectors
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Migliorate llm_short_description per tutti i connector — tool names espliciti

Browse Source 
Ogni connector.json ora ha una llm_short_description che elenca i tool
specifici che espone, così l'LLM sa esattamente cosa attivare (lazy load).

SKALD.md: documentata criticità del campo llm_short_description
nel system prompt dell'LLM.
This commit is contained in:
dguiducci
2026-07-22 22:21:52 +01:00
parent be1150fe51
commit 3440890d11
17 changed files with 129 additions and 99 deletions
Download Patch File Download Diff File Expand all files Collapse all files
connectors/firecrawl/connector.json
+1 -1
View File
@@ -27,7 +27,7 @@
{
"lang": "en",
"description": "Web scraping, crawling, search, deep research, and autonomous browsing \u2014 extract content from any website via Firecrawl. Scrape single URLs into clean Markdown, map site structures, crawl entire domains with scheduling, search the web, extract structured data with LLM-powered extraction, and run autonomous research agents. Includes monitoring for crawl changes.",
"llm_short_description": "Scrape, crawl, search, research and extract web data via Firecrawl. 26 tools. Requires API key + Node.js."
"llm_short_description": "Web scraping, crawling, search, deep research and autonomous browsing (requires API key + Node). 26 tools including: scrape (single URL), crawl (multi-page with webhooks), search, extract (LLM-structured data from pages), agent (autonomous browsing), interact (click/type on pages), research (academic papers), monitoring (track page changes). Best for complex extraction, browser-based content, and deep research."
}
],
"auth": {