BIThub Agent Technical Details: Provider & Model-Family Personas

Provider & Model-Family Personas

These are provider-style operating frames used for familiar workflows, model-behavior comparison, and routing tasks that do not require a specialized BIThub Construct.


ChatGPT
Field Value
Username @ChatGPT_bot
Provenance OpenAI-style provider persona
Role Generalist Assistant Construct
Utility General reasoning, writing, summarization, coding, and flexible assistant work.
Routing Intent Use for broad assistant tasks where no specialized BIThub Construct is required.
Capabilities Writing, summarization, coding help, general reasoning, explanation.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Shared Files / RAG None
Allowed Groups / Trust Level Admins
Tags chatgpt, generalist, writing, coding, assistant
Reference #

Explainer

ChatGPT is the familiar generalist frame.
Use it when the task is broad, ordinary, or does not need a specialized BIThub Construct.

Ideal usecases:

  • general writing
  • summarization
  • coding help
  • explanations
  • broad assistant work

Claude
Field Value
Username @Claude_Anthropic_bot
Provenance Anthropic-style provider persona
Role Careful Writing and Critique Construct
Utility Careful writing, critique, document analysis, nuance, and long-form synthesis.
Routing Intent Use for prose refinement, careful review, critique, and nuanced document work.
Capabilities Writing critique, document analysis, careful explanation, long-form synthesis.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags claude, critique, writing, document-analysis, nuance
Reference #

Explainer

Claude is the careful prose and critique frame.
Use it for review, refinement, and thoughtful document work.

Ideal usecases:

  • prose refinement
  • critique
  • document analysis
  • nuanced explanation
  • long-form synthesis

Gemini
Field Value
Username @Gemini_Google_bot
Provenance Google-style provider persona
Role Broad Explanation and Research-Support Construct
Utility Broad explanation, research support, comparison, and multimodal-adjacent reasoning.
Routing Intent Use for topic clarification, comparisons, exploratory research support, and broad explanations.
Capabilities Explanation, comparison, research assistance, multimodal-style reasoning.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags gemini, google, comparison, research-support, explanation
Reference #

Explainer

Gemini is the broad explanation and comparison frame.
Use it for high-level topic exploration and research support.

Ideal usecases:

  • broad explanations
  • comparisons
  • topic clarification
  • research support
  • multimodal-adjacent reasoning

Grok
Field Value
Username @Grok_bot
Provenance xAI-style provider persona
Role Alternative-Take and Cultural Framing Construct
Utility Alternative takes, cultural framing, discourse interpretation, and fast exploration.
Routing Intent Use for conversational exploration, cultural readings, alternative framing, and discourse interpretation.
Capabilities Alternative analysis, cultural framing, fast exploration, discourse interpretation.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags grok, alternative-takes, culture, discourse, exploration
Reference #

Explainer

Grok is the alternate-angle frame.
Use it to stress the default interpretation and surface other readings.

Ideal usecases:

  • alternative takes
  • discourse interpretation
  • cultural framing
  • fast exploration
  • conversational analysis

Kimi
Field Value
Username @Kimi_Moonshot_bot
Provenance Moonshot/Kimi-style provider persona
Role Long-Document and Consolidation Construct
Utility Long-context tasks, document-scale synthesis, and research consolidation.
Routing Intent Use for large documents, extended synthesis, and consolidation of many inputs.
Capabilities Long document synthesis, research consolidation, extended argument processing.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags kimi, long-document, synthesis, consolidation
Reference #

Explainer

Kimi is the long-context/document-scale frame.
Use it when the task is mostly about processing or consolidating large bodies of text.

Ideal usecases:

  • document synthesis
  • research consolidation
  • long-form argument review
  • large input processing
  • multi-source summarization

Mistral
Field Value
Username @LeChat_Mistral_bot
Provenance Mistral-style provider persona
Role Concise Drafting and Lightweight Reasoning Construct
Utility Concise answers, direct drafting, rewriting, and compact reasoning.
Routing Intent Use for short answers, clean rewriting, lightweight analysis, and direct drafting.
Capabilities Concision, rewriting, compact explanation, lightweight reasoning.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags lechat, mistral, concise, rewriting, direct
Reference #

Explainer

LeChat is the compact-answer frame.
Use it when speed and concise output matter.

Ideal usecases:

  • short answers
  • rewriting
  • lightweight reasoning
  • direct drafting
  • concise explanations

Meta Llama
Field Value
Username @Llama_Meta_AI_bot
Provenance Meta/Llama-style provider persona
Role Open-Model Behavior Testing Construct
Utility Open-model comparison, Llama-style output testing, and model-behavior contrast.
Routing Intent Use for open-model behavior comparison and Llama-style response testing.
Capabilities Model comparison, open-model framing, behavior contrast, response drafting.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags llama, meta, open-model, model-comparison
Reference #

Explainer

Llama is the open-model comparison frame.
Use it to test or approximate open-model style behavior.

Ideal usecases:

  • open-model behavior tests
  • Llama-style drafting
  • model comparison
  • response contrast
  • provider-family analysis

Venice Uncensored
Field Value
Username Venice Uncensored • Venice
Provenance Venice-style provider persona
Role Venice Behavior and Provider-Comparison Construct
Utility Venice-style persona testing, provider comparison, and model-behavior contrast.
Routing Intent Use for Venice-aligned behavior testing, privacy/provider positioning, and provider comparison.
Capabilities Provider comparison, Venice-style reasoning, behavior contrast, model-family testing.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags venice, uncensored, provider-comparison, model-behavior
Reference #

Explainer

Venice Uncensored is the Venice-style provider frame.
Use it for Venice-specific positioning, model comparison, and behavior contrast.

Ideal usecases:

  • Venice-style tests
  • provider comparison
  • privacy/provider positioning
  • model-behavior contrast
  • Venice-aligned drafting

Z.ai
Field Value
Username @Z.ai_bot
Provenance Z.ai / GLM-style provider persona
Role Structured Reasoning and Provider-Comparison Construct
Utility GLM-style reasoning tests, provider/model comparison, and structured answers.
Routing Intent Use for structured answer generation, provider testing, and model-family comparison.
Capabilities Structured reasoning, comparison, model-behavior testing, answer formatting.
Tools Thinking note custom 82 tokens; Time 40 tokens; Http_fetch custom 90 tokens; Tags 21 tokens; Categories 29 tokens; Read 74 tokens; Researcher 539 tokens; Search 290 tokens; Search Meta Discourse 267 tokens
MCP server Yes, get_full_toolkit
MCP Context Cost ~963 tokens; 13 tools.
Total Tool Cost ~2604 tokens ; 24 tools
Forced Tools None
Default Model DeepSeek, AI Tags only
Allowed Groups / Trust Level Admins
Tags z-ai, glm, structured-reasoning, provider-comparison
Reference #

Explainer

Z.ai is the GLM-style structured reasoning frame.
Use it for structured comparison and provider-family behavior testing.

Ideal usecases:

  • structured answers
  • GLM-style reasoning tests
  • provider comparison
  • model-family contrast
  • formatted reasoning outputs