grok_reasonDeep first-principles reasoning with optional heavy mode.Grok (xAI)
openai_reasonMathematical & high-effort reasoning with GPT-6 Astra at xhigh reasoning effort.OpenAI
qwen_reasonHeavy mathematical reasoning (Qwen3.8 Max, 1M context).Qwen
qwq_reasonMulti-perspective deliberation: optimist, pessimist, expert, contrarian.Qwen
kimi_thinkingMultimodal reasoning with a 100-subagent swarm.Kimi
perplexity_reasonReasoning grounded in live web search.Perplexity
deepseek_reasonOpen-weight frontier reasoning — top AIME/GPQA math & logic chains.DeepSeek
glm_reasonAgentic reasoning & tool-use planning (SWE-Bench Pro leader).GLM (Z.ai)
stepfun_reasonEfficient deep reasoning — high AIME/SWE-Verified at flash-tier cost.StepFun
ernie_reasonBroad-knowledge reasoning with human-preference/arena strength.ERNIE (Baidu)
local_queryQuery a local model via Ollama/LM Studio/llama.cpp/vLLM — free, offline, private.Local
nextThoughtSequential thinking chain with optional per-step model execution.TachiBot
thinkLightweight internal reasoning log.TachiBot
qwen_coderAgentic code generation. SWE-bench >70%.Qwen
qwen_algoAlgorithm analysis: complexity & optimization tiers.Qwen
deepseek_algoStrongest algorithmic review: correctness, Big-O, edge cases, data structures.DeepSeek
qwen_competitiveCompetitive-programming problem solving.Qwen
kimi_codeSWE-focused code generation & bug-fixing leader.Kimi
minimax_codeSingle-pass code operations, cost-efficient (1M context).MiniMax
grok_codeCode analysis & optimization.Grok (xAI)
grok_debugSpecialized debugging assistance.Grok (xAI)
debug_triageRanked root-cause hypotheses with likelihoods, a discriminating check per hypothesis, and the minimal fix for the leader.Grok (xAI)
openai_code_reviewComprehensive code review with focus areas.OpenAI
testgenGenerate runnable tests via Qwen3-Coder-Next — edge cases enumerated first.Qwen
diff_reviewMulti-model diff-aware review — Kimi K3 + DeepSeek + GPT-6 Astra scoped to changed lines, Gemini-judged verdict.TachiBot
perplexity_askWeb search with current information (Sonar).Perplexity
grok_searchLive web search with Grok 4.6.Grok (xAI)
grok_search_liteCheaper live search on grok-4.3 — high-volume lookups and jury fan-outs.Grok (xAI)
openai_searchReal-time web search via GPT-5.6.OpenAI
gemini_searchWeb search with Google Search grounding.Gemini
gemini_analyze_codeCode quality, security & performance analysis.Gemini
gemini_analyze_textRhetorical analysis: bias, fallacies, persuasion.Gemini
kimi_long_contextLong-context analysis across a 1M window.Kimi
openai_explainClear explanations for complex topics, any level.OpenAI
security_reviewOWASP/CWE security audit via DeepSeek V4 Pro — taint/data-flow analysis, severity per finding, concrete fixes.DeepSeek
grok_brainstormContrarian first-principles idea generation.Grok (xAI)
openai_brainstormFind the 3rd, 4th & 5th alternative approaches.OpenAI
gemini_brainstormConvergent synthesis — clusters ideas into themes.Gemini
juryMulti-model jury panel → Gemini judge synthesis.TachiBot
gemini_judgeMulti-perspective evaluation & verdict synthesis.Gemini
planner_makerCouncil-based plan creation via coordinator pattern.TachiBot
planner_runnerExecute plans against goal-oriented verification gates.TachiBot
list_plansList recently created plans.TachiBot
kimi_decomposeStructured task decomposition via agent swarm.Kimi
minimax_agentMulti-step task decomposition & execution.MiniMax
grok_architectSystem architecture & design via 4–16 agent swarm.Grok (xAI)
plan_critiqueAdversarial multi-model plan red-team — pre-mortem, hidden assumptions, ranked risks, Gemini-judged verdict.TachiBot
spec_writerLoose request → reviewable spec via GPT-6 Astra: user stories, Given/When/Then, out-of-scope, open questions. Sign off before planning.OpenAI
blog_writerResearched long-form in one call — persona interview for the angle, then plan/draft/restructure/line-edit passes (storm + re3).Kimi
workflowExecute YAML-based multi-step workflows.TachiBot
workflow_startBegin a workflow session with variables.TachiBot
continue_workflowResume an interrupted workflow session.TachiBot
list_workflowsDiscover available workflows.TachiBot
create_workflowCreate a new YAML workflow definition.TachiBot
workflow_statusCheck progress of a running workflow session.TachiBot
validate_workflowValidate workflow YAML: syntax, tool registry, dependency graph.TachiBot
validate_workflow_fileValidate a workflow file on disk before running.TachiBot
visualize_workflowRender a workflow's step graph for inspection.TachiBot
list_prompt_techniquesDiscover techniques — ~9 core by default, all=true for the full 31.TachiBot
preview_prompt_techniquePreview an enhanced prompt before executing, or recommend techniques with technique="auto".TachiBot
execute_prompt_techniqueRun a query with a chosen technique applied.TachiBot
refine_promptOpt-in prompt improver — raw query → goal-first brief + what-changed + open questions. Never executes.OpenAI
tachiSmart auto-router: research, solve, verify, create, judge.TachiBot
focusMode-based multi-model reasoning (deep, debate, analyze).TachiBot
continue_focusContinue a focus session by its session ID.TachiBot
usage_statsView or reset tool usage & cost statistics.TachiBot
doctorSetup diagnostic — detected API keys, visible vs hidden tools (and why), active profile, suggested first step. Zero-cost.TachiBot