Email Extraction
FreeFeeding a whole 50-message thread to a model burns tokens on quoted history; this returns just the new reply, with offsets back to the original.
Documents
Strip quoted history and reply chains from an email, returning just the new or relevant content as a verbatim slice of the input, plus a structured breakdown (new, quoted, attribution) with offsets back into the original. Stateless and extractive: nothing is rewritten, summarized, or retained. Free to use, with fair-use rate limits.
Use when
You have a raw email and want only the new reply, e.g. to cut token cost before feeding it to a model.
Not for
Generating or summarizing email, parsing attachments, or fetching mail from a provider (you supply the body).
What you can do
Each is a sub-skill of Email Extraction; the router picks the right one for your request.
What you get back
information
Returns the cleaned body verbatim plus the method, a truncated flag, a version string, and new/quoted/attribution offsets into the input. Extractive only — never contains text that was not in the input. Not a downloadable file.
When it checks with you first
Neither an HTML nor a text body is provided.
Run it
Skills run through one gateway with your Faro token. Hand it an intent in plain language; Faro routes to the right sub-skill, runs it, and bills per call.
curl -X POST "https://skill.askfaro.com/skills/email-extraction/run" \
-H "Authorization: Bearer $FARO_TOKEN" \
-H "Content-Type: application/json" \
-d '{"intent":{"prompt":"Strip the quoted history from this email and return just the new reply."}}'Example requests
- ›Strip the quoted history from this email and return just the new reply.