Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
analyze-chats - Prompt - OpenSmartRoute
Promptv1.0.0
analyze-chats
Analyze each and every chat log. - After each file, identify instances where I(the developer) corrected or clarified the model's behavior.
Prompt file imported from anoblet/genaisrc (.github/prompts/analyze-chats.prompt.md). Copyright stays with the author.
Analyze each and every chat log.
After each file, identify instances where I(the developer) corrected or clarified the model's behavior.
Do not record the fact I told the model yes, continue, or to keep going.
Record these observations to memory.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
- Analyze each and every chat log.
- After each file, identify instances where I(the developer) corrected or clarified the model's behavior.
- Do not record the fact I told the model yes, continue, or to keep going.
- Record these observations to memory.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "- Analyze each and every chat log.\n- After each file, identify instances where I(the developer) corrected or clarified the model's behavior.\n- Do not record the fact I told the model yes, continue, or to keep going.\n- Record these observations to memory.",
"variables": [],
"notes": "Analyze each and every chat log. - After each file, identify instances where I(the developer) corrected or clarified the model's behavior. - Do not record the fact I told the model yes, continue, or to keep going. - Record these observations to memory.",
"metadata": {
"source": {
"provider": "github-prompts",
"repository": "https://github.com/anoblet/genaisrc",
"path": ".github/prompts/analyze-chats.prompt.md",
"ref": "21a8d6630688dc9fcf17f6197a7d830dd75713c5",
"url": "https://github.com/anoblet/genaisrc/blob/21a8d6630688dc9fcf17f6197a7d830dd75713c5/.github/prompts/analyze-chats.prompt.md",
"key": "anoblet/genaisrc/.github/prompts/analyze-chats.prompt.md"
}
}
}
Installed into a catalogue, chosen by a router
Install analyze-chats and it becomes one more candidate the router can pick - when it fits.
A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.