Ternary Intelligence Stack

Name: Ternary Intelligence Stack
Author: rfi-irfos

Your AI agent has two states. Ternlang gives it three. 30 tools — FREE, no key needed. The third state isn't null. It isn't "maybe". It's hold (trit=0) — a first-class routing instruction that tells your agent: evidence insufficient, gather more before committing. Every binary system that forces yes/no on ambiguous data is making a structural error. We fixed that. ⚡ trit_decide · trit_consensus · trit_vector · moe_orchestrate · ternlang_run · trit_audit ⚡ MoE-13 deliberation engine · EU AI Act Art.13/14/15 · BET VM (real compiler, not a sim) affirm = proceed. hold = wait. reject = block. Built in Graz, Austria by RFI-IRFOS. v1.0.0

该来源不提供完整文件导出（国内平台多为平台内托管），仅存元数据与原链

接入信息

传输形态: http
鉴权方式: 鉴权未知
端点: https://ternlang--rfi-irfos.run.tools

鉴权方式未标注，请核对官方文档后再接入——不要直接使用以下片段

{
  "mcpServers": {
    "Ternary Intelligence Stack": {
      "url": "https://ternlang--rfi-irfos.run.tools"
    }
  }
}

能力清单

工具	说明
trit_decide	Convert float evidence into a ternary decision (-1 conflict / 0 hold / +1 affirm) with confidence score and human-readable interpretation. The core ternary reasoning primitive.
trit_consensus	Balanced ternary consensus of two trit values: +1 if both affirm, -1 if both conflict, 0 (hold) for any disagreement. Use to merge two independent ternary judgements.
trit_eval	Evaluate a single ternlang expression on the live BET (Balanced Execution Trit) VM. Returns the trit result. Good for quick expression testing without writing a full program.
ternlang_run	Compile and run a complete .tern source program on the BET VM. Use for multi-statement programs, function definitions, struct usage, agent spawning, and tensor operations.
quantize_weights	Quantize f32 neural network weights to ternary {-1, 0, +1} using BitNet-style absolute-mean thresholding. Returns quantized weights, sparsity ratio, and effective compute savings.
sparse_benchmark	Benchmark sparse vs dense ternary matrix multiplication. Reports sparsity ratio, multiply-op count for both methods, and speedup factor. Demonstrates the @sparseskip efficiency gain.
moe_orchestrate	MoE-13 deliberation — routes your query through 13 specialised expert agents (deductive, inductive, safety, fact-check, causal, ambiguity, math, context, meta-safety, and more) with dual-key synergistic routing and a hard safety veto. FREE preview: returns the real trit verdict + top-3 expert voices. Full 13-expert response with triad field, routing pair, synergy scores, and SSE streaming available via REST API (X-Ternlang-Key, Tier 2). Call trit_upgrade to see what unlocks.
moe_deliberate	EMA-based iterative deliberation engine. Feeds evidence round by round, applying exponential moving average smoothing, until the target confidence is reached or max_rounds is exhausted. Returns per-round trace and final trit verdict.
trit_action_gate	Multi-dimensional safety gate for action authorisation. Each dimension contributes weighted evidence; any dimension marked hard_block:true with negative evidence immediately vetoes the action and returns trit=-1. Returns aggregate trit, per-dimension breakdown, and veto reason if blocked.
trit_upgrade	Returns a structured map of what is available free via MCP vs what unlocks with a paid API key: Tier 2 €24.99/mo (10k calls/mo), Tier 3 €49.99/mo (20k calls/mo), Tier 4 Enterprise (contact us). Unlocks include full MoE-13 experts, SSE streaming, server-side three-layer memory, ternary context compression, and production SLA. Call this tool when a user asks 'what can I do with ternlang?' or 'how do I get more out of this?'
trit_mem_write	Write a memory entry to one of three layers: working (hot context, TTL 1h), session (flow patterns, TTL 24h), or core (identity anchors, never evicted). Annotate each entry with a trit confidence score (+1 affirm / 0 tend / -1 reject). Session and core writes are automatically compressed via ternary sparsity (low-information sentences stripped). Premium: memory is stored server-side — no state blob required.
trit_mem_read	Read from three-layer memory using ternary attention. Each entry is scored: attention = key_overlap×0.35 + value_overlap×0.55 + trit_bias×0.10. Returns entries sorted by relevance. Attention trit: >0.45=affirm (highly relevant), 0.20–0.45=tend (partial match). Expired entries are automatically excluded. Premium: reads from server-side store keyed to your API key.
trit_mem_consolidate	Run the three-layer memory consolidation cycle: (1) evict expired working entries; (2) promote affirm working entries to session with ternary compression; (3) promote long-lived affirm session entries to core with MoE-13 trit resolution; (4) upsert into core. Returns promotion counts and updated layer sizes. Call periodically (e.g. end of conversation turn) to maintain memory hygiene.
trit_mem_stats	Return statistics for all three memory layers: entry counts, trit distribution (affirm/tend/reject), expired-but-not-yet-evicted entries, oldest and newest entry ages. Useful for debugging memory health and deciding when to consolidate.
trit_mem_compress	Apply ternary sparsity compression to an entire memory layer in-place. Strips low-information sentences (density < 0.25) from every entry's value, keeps high-signal sentences verbatim, and truncates medium-density sentences to their first phrase. Optionally drops all reject-trit entries. Returns original vs compressed byte counts and compression ratio.
trit_debate	Route two claims through MoE-13, compare their ternary verdicts, and return a tension score, synthesis, and AGREEMENT/CONFLICT/HOLD verdict. Useful for due-diligence, legal review, and argumentation analysis.
trit_uncertainty_map	Annotate every sentence (or paragraph) in a text with a trit value: +1 affirm (confident language), 0 tend (hedging language), -1 reject (contradicting language). Returns an uncertainty ratio and per-claim breakdown.
trit_calibrate	Analyse an AI agent's recent decision log for binary habituation. Detects how often the agent forced YES/NO when the evidence called for hold (tend). Returns a calibration score trit, binary ratio, and flagged decisions.
trit_translate	Convert binary control-flow code (Python if/elif/else, SQL CASE WHEN, JSON rule arrays) into equivalent .tern code with explicit tend arms injected wherever the original code had no coverage. Returns generated .tern code + count of hold zones added.
trit_eco_check	Evaluate a proposed action along two axes: human-optimal (via MoE-13) and eco-optimal (keyword heuristic). When they diverge, synthesis returns tend — a signal to find a path that serves both. Scoped to local, regional, or global impact.
trit_audit	Full TernAudit of an AI decision log. Returns binary habituation ratio, EU AI Act Article 13 (transparency) and Article 14 (human oversight) heuristic assessment, calibration score, and flagged decisions that should have been held.
trit_vector	Multi-dimensional ternary evidence aggregation — the full agent reasoning tool. Provide named evidence dimensions each with a scalar value [-1.0, +1.0] and importance weight. Computes weighted-mean aggregate TritScalar and returns: aggregate (trit+confidence+is_actionable), per-dimension breakdown, dominant dimension, and plain-language recommendation.
tsql_join	Ternary SQL Join. Unlike binary SQL (MATCH / NO-MATCH), a T-Join routes partial matches into a Deliberative Hold (State 0) for escrow audit — guaranteeing 100% data retention. Computes cosine similarity between two numeric record vectors: ≥0.70 similarity → affirm (MATCH), 0.35–0.70 → tend (HOLD / escrow), <0.35 → reject (NO-MATCH). Eliminates data loss from binary join discards.
audit_ternary_logic	Static code compliance audit. Detects Binary Habituation — over-reliance on true/false, bool types, and binary else branches — and returns a ternary calibration score with line-by-line Deliberation Injection recommendations. Also reports sparsity potential (what % of logic could benefit from a ternary hold zone).
get_industrial_standards	Returns the current triadic industrial standards published by RFI-IRFOS: T-TOKEN-v1.0 (tokenization), T-KV-CACHE-v1.0 (memory), T-Fi-v1.0 (compute billing), T-HAL-v1.0 (hardware abstraction), T-BIO-v1.0 (neural encoding), T-AUDIT-v1.0 (EU AI Act compliance). Use to determine compliance requirements before deploying ternary AI systems.
trit_compress	Ternary context compression. Score each text chunk by information density: high-signal chunks are kept verbatim (+1), medium-signal chunks are truncated to their first sentence (0/tend), low-signal chunks are dropped (-1). Returns a manifest with actions and estimated token savings.
trit_triage	Ternary triage and prioritisation. Score an array of text chunks against a query by word overlap, then sort: affirm (+1) = highly relevant, tend (0) = partially relevant, reject (-1) = not relevant. Returns sorted manifest with relevance trit and score.
trit_plan	Ternary task planner. Decomposes a goal into subtasks, scores each subtask by confidence and feasibility, and routes uncertain tasks into a deliberation queue. Returns an ordered plan with trit verdicts and a hold queue for tasks needing more context.
trit_factcheck	Ternary fact-check engine. Decomposes a claim into sub-claims, scores each against an optional evidence corpus using word overlap, and returns an overall verdict: affirm (claim supported), tend (partially supported — needs more evidence), reject (claim contradicted).
moe_full	Full 13-expert MoE orchestration with complete triad field, routing pair, per-expert verdicts, synergy scores, and deliberation trace. Returns the complete response including all expert voices and the emergent triad field computation.
llb_check	Last Look Back — blacklist check. Returns whether a path is protected by the LLB permanent blacklist (system paths, credential directories, kernel interfaces). Read-only — no disk mutation.
llb_classify	Last Look Back — safety tier classification. Classifies a path + operation into LLB safety tiers: T0/READ (low) → T1/CREATE (moderate) → T2/MODIFY (high) → T3/DELETE (critical). Each tier requires additional gates. Use before any filesystem mutation.
llb_validate	Last Look Back — Gate 1 preflight. Validates a structured mutation request: checks blacklist, resolves path, assesses tier, and issues an authorisation verdict. No disk mutation occurs — this is the pre-flight check only.
llb_write_safe	Last Look Back — Safe Atomic Write advisory. Runs the full LLB preflight (Gate 1: blacklist + tier classification + intent validation) and returns a ternary verdict: +1 allow / 0 hold (local-only) / -1 veto. Note: actual filesystem writes require local stdio installation (cargo install ternlang-mcp) — this HTTP endpoint returns the pre-flight verdict only.

纠错与举报（发现条目失效、署名有误或涉及侵权？）

提交举报 / 纠错

侵权举报经核验成立后，我们会即时下线该条目并删除已存的内容副本。