LotusGuild/matrix
2
0
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases Wiki Activity

ask: instruct LLM not to ask follow-up questions
Lint / Shell (shellcheck) (push) Successful in 9s
Details
Lint / JS (eslint) (push) Successful in 7s
Details
Lint / Python (ruff) (push) Successful in 5s
Details
Lint / Python deps (pip-audit) (push) Successful in 1m6s
Details
Lint / Secret scan (gitleaks) (push) Successful in 5s
Details

Browse Source 
Each !ask call is stateless — no context is retained between commands,
so ending a response with a question is misleading. Added explicit
instruction to the system prompt to prevent this.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Jared Vititoe
2026-04-20 17:10:26 -04:00
parent 637b2a4b20
commit e14b9a274f
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
matrixbot/commands.py
+2
View File
@@ -805,6 +805,8 @@ async def cmd_ask(client: AsyncClient, room_id: str, sender: str, args: str):
"Answer questions clearly and concisely. Keep responses reasonably brief — " "Answer questions clearly and concisely. Keep responses reasonably brief — "
"a few sentences to a short paragraph unless the question genuinely needs more detail. " "a few sentences to a short paragraph unless the question genuinely needs more detail. "
"Be friendly and conversational. " "Be friendly and conversational. "
"Do NOT ask follow-up questions or prompt the user to continue — "
"each message is standalone with no conversation history."
), ),
}, },
{"role": "user", "content": question}, {"role": "user", "content": question},