arifOS
by ariffazil
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GEOX_BRIDGE_HOST | No | Override host for GEOX bridge. | http://127.0.0.1:8081 |
| WELL_BRIDGE_HOST | No | Override host for WELL bridge. | http://127.0.0.1:18083 |
| AAA_MCP_TRANSPORT | No | Transport mode for MCP server (stdio or sse). Default is auto-detected. | |
| WEALTH_BRIDGE_HOST | No | Override host for WEALTH bridge. | http://127.0.0.1:18082 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
| resources | {
"subscribe": true,
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| arif_initD | arif_init |
| arif_observeD | arif_observe |
| arif_thinkD | arif_think |
| arif_judgeD | arif_judge |
| arif_memoryD | arif_memory |
| arif_routeD | arif_route |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| 000 π± IGNITE | Position Zero (/000) Sovereign Anchor. Identity before action. REALITY > EVERYTHING. |
| 111 π SENSE | Reality before judgment. Verify before integrate. |
| 222 π PLAN | Design reality change. Map reversibility. |
| 333 π§ REASON | UNMEASURED beats fabricated certainty. |
| 444 π§ DIRECT | Route to the institution with authority, not the one with speed. |
| 555 π REMEMBER | Memory without provenance is not truth. |
| 666 β DIGNITY | Stand in the position of the weakest stakeholder. |
| 777 π₯ FORGE | Reality contact before belief. Mutation after SEAL only. |
| 888 π JUDGE | Verdict, not invention. VOID = branch dead. |
| 999 π SEAL | Immutable record. Hash-chained. Companion chamber to /000. Cannot be undone. |
| π GOVERN | Full loop + 4 invariant enforcement gates. GATE 1 GΓDEL LOCK: every loop touches reality before becoming doctrine. GATE 2 ANTI-SINK: signal β verify β reality β integrate (no amplified garbage). GATE 3 ENTROPY: ΞS β€ 0 per cycle. GATE 4 VOID: dead branches stay dead without new evidence. |
| β INIT | Collapsed 4-step governed ignition (000_INIT). Probes kernel, binds session, loads context. |
| π CLOSE | Full autonomous session close ritual (999_CLOSE). Two-lane seal/receipt. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| get_vitals | Real-time constitutional vitals and thermodynamic telemetry. Returns CPU, memory, disk, genius score (G), entropy delta (ΞS), human impact load (Ξ©), and paradox tension (Ξ¨). Updated continuously by the metabolic monitor. Use for health checks. |
| agent_init | INIT.md v4.0 SUPERSEDED 2026-09-04 β file is now a redirect stub (MCP /init). Canonical init = arif_init (KERNEL 000) + command/init.md v7.0. Legacy 293-line content preserved at _archive/2026-09-16-prompt-zen/. Fix 2026-10-18 (FI-008): description drift β was advertising 293 lines against an 8-line stub. |
| get_carry_forward | Live session carry-forward state (v3 typed entries). Returns prior session decisions, scars, open loops, eurekas, directives, and anchors. Schema-gated: only serves arifos.carry_forward.v3. Essential for agent continuity β load at session start. |
| get_flow_state | Live Flow Quality (FQ) pulse β the federation's metabolic nerve health. Returns FQ value, verdict (OPTIMAL/BALANCED/WATCHING/STUCK), and last update. FQ < 0.5 β ALL agents HOLD (OBSERVE_ONLY). FQ >= 0.5 β forge. This replaces filesystem reads of /root/AAA/state/flow_state.json. Cross-reference with arifFlow :7073/health for real-time metabolic data. |
| get_instructions | Agent bootstrap instructions β the MCP-native equivalent of server/discover. Every agent entering arifOS for the first time should read this. Contains boot sequence, key resources, canonical tools, and constitutional rules. |
| Constitutional Doctrine | The immutable 13-floor constitution (F1βF13) that governs all arifOS operations. Includes Amanah, Truth, Witness, Clarity, Peace, Maruah, Humility, Genius, Anti-Hantu, Ontology, Audit, Injection, and Sovereign. Source of truth for floor definitions. Audited 2026-07-15. |
| arifOS Namespace Index | Canonical resource directory β counts computed live from the registry at read time. |
| Federation Skill Index | Federation skill directory β counts computed live. Use skill://{name}/SKILL.md. |
| Refusal Surface | Complete catalogue of what the arifOS kernel refuses and why. Hard refusals (VOID), soft refusals (888_HOLD), conditional refusals (HOLD), and epistemic refusals. The negative space of the constitution. |
| AAA Trinity Lanes | The AAA Trinity lane architecture: AGI proposes (111), ASI judges (444), APEX authorizes (888), FORGE executes (010), 999 seals. Defines which tools belong to which lane and which domain language each lane uses. SEAL/SABAR/HOLD/VOID are APEX-only verdict terms. |
| Seal Readiness & Vault Integrity | Vault integrity report and seal gate requirements. Defines the five disambiguated seal types, the seven-floor seal gate, and the non-seal verdicts (SABAR/HOLD/VOID). Bare SEAL without namespace is non-compliant. |
| Institutional Ontology | Seven federation organs as institutions (decides, executes, witnesses, computes, reflects, operates, judges), four constitutional lanes (AGI proposes, ASI judges, APEX authorizes, FORGE executes), five autonomy bands (GREEN to SOVEREIGN), and the definition of institution: authority boundary, distinct context, accountability. |
| Canonical Schema | Complete canonical blueprint of the arifOS MCP surface. Lists 8 canonical tools (capability-registered, wire-verified), 13 constitutional floors (F1βF13), and the separation of powers doctrine. Source of truth: abi/capability_registry.json. Audited 2026-07-15. |
| Civilizational Ontology | The seven federation organs (arifOS, A-FORGE, GEOX, WEALTH, WELL, AAA, APEX), three intelligence strata (AGI, ASI, APEX), constitutional boundaries between domain organs and the kernel, and entropy responsibility model. Defines which organs DECIDE vs WITNESS vs COMPUTE vs REFLECT. |
| Jurisdiction & Autonomy Bands | The five autonomy bands (GREEN through BLACK) that govern agent freedom, the CapabilityGrant registry for secretless execution, and jurisdiction rules. Agents hold grants, never raw secrets. Autonomy is a leased capability, not a permanent possession. All tools mapped to their autonomy band. |
| Sovereign Identity Manifest | Sovereign identity manifest bound from identity.toml at boot. Defines the authority chain from APEX (Arif) through arifOS kernel to domain organs and execution. Includes ADR-001 localhost doctrine. Identity is the root of accountability β all attestation chains begin here. |
| Memory Architecture (L1βL6) | The six-layer memory architecture: L1 ephemeral (Redis), L2 session (Redis), L3 semantic (Qdrant/BGE-M3), L4 structured (Supabase), L5 retired (Graphiti, β L3/L4), L6 immutable (VAULT999). Memory does not become truth until it has provenance. Truth does not become final until sealed. |
| Federation Bootstrap Context | Complete arifOS federation knowledge-graph bootstrap context. Delivers the full world model: organ identities, agent roster, current runtime state, constitutional law (F1-F13), ROOTKEY/EUREKA modules, known open loops, safe defaults, tasking template, and epistemic style guide. Fetch this when you need the complete federation map in one call. v2026.06.14. |
| 7-Stage Reality Engineering Loop | Documents the arifOS 7-stage reality engineering loop: arifosmcp_loop_engineer (entry guard) + 000_init/111_sense/333_reason/555_judge/666_critique/777_forge/999_seal. Each stage has dual naming (K1: numeric internal ID + function name). Includes organ routing table, reversibility classification, blast radius, evidence truth hierarchy, and uncertainty substrate mapping. MCP prompt surface is at arifosmcp_prompts module. |
| LLM Client Quickstart | Getting started guide for LLM clients connecting to arifOS MCP. Covers: endpoint, transport, auth, canonical tools, resources, prompts (counts live in arifos://index β single SOT, never hardcoded here), 7 stages, key constitutional rules (F1-F13), when to use arifOS vs direct organ, and VAULT999 ledger guidance. Read this first before using the server. |
| get_tree777_index | TREE777 wiki full index. Lists all skills (by category), concepts, and scars in the canonical wiki. Use this to discover what resources are available before fetching individual pages. |
| sovereign_000_genesis_resource | arifOS Position Zero Genesis Statement (/000) signed by Muhammad Arif bin Fazil. |
| sovereign_000_claims_resource | arifOS Position Zero Attestation Claims (000-CLAIM-001..006, ZKPC dimensions, GΓΆdel Lock). |
| human_metabolized | Metabolized sovereign context β compact human intelligence. Returns the structured signal that changes how AI behaves around this human. Not the full testimony β the nutrient. Source: AAA/wiki/SCAR_TERRAIN.md (SOVEREIGN_TESTIMONY; renamed from scar-terrain-arif-fazil.md) Derived: 2026-06-16, FORGE (000Ξ©) |
| reality_state_resource | Causally-consistent multi-layer reality snapshot. Returns current state of PHYSICAL, DIGITAL, BIOLOGICAL, and UNCERTAINTY_SUBSTRATE reality layers at one consistent timestamp. Every value carries epistemic label (OBS/DER/INT/SPEC/UNKNOWN). Uncertainty substrate maps quantum-physics metaphor to classical decision engineering. |
| wisdom_quotes_all | Return all quotes from the canonical registry with provenance metadata. |
| wisdom_quotes_disputed | Quotes with DISPUTED_ATTRIBUTION β shadow watch, never load-bearing. |
| wisdom_quotes_arifos_doctrine | Native arifOS doctrine entries (ratification ladder applies). |
| wisdom_quotes_prohibited_uses | Prohibited-use index β governance boundary per quote. |
| wisdom_contract | Federation contract for the wisdom/quote namespace. Single source of truth for cross-organ integration. Defines: - 7 APEX organs + multiplicative G formula - shadow_state taxonomy (Pillar VI) - canon_status tier ladder - stage binding (555_HEART | 999_RECEIPT only) - citation rules (prohibited: factual_evidence, verdict_authority) |
| skill_health_resource | Catalog-wide skill health scan. Returns total/complete/stub/stale counts, health percentage, broken file references, and ecology state (HOT/WARM/COLD) from Qdrant. Dynamic β computed from live filesystem on each read. |
| epistemic_resource | Epistemic label ontology (OBS/DER/INT/SPEC + provenance rules). |
| floors_resource | All 13 constitutional floors as a reference list. |
| affordances_resource | Tool affordances: action classes, blast radius, reversibility. |
| vault_head_resource | VAULT999 chain head β live from filesystem. |
| atlas333_index | ATLAS333 root index β consolidated cognitive geometry. |
| atlas333_metrics | Institutional MAPΒ·ATLASΒ·ECHO as ATLAS333 geometry input. Wiring layer only (F2/F8): MAP β paradox selection density (top_k calibration) ECHO β live tension weights (P2 heats on memory visibility) ATLAS metric β judge-axis heat when governance compression low ATLAS333 content (35 paradoxes) is immutable compression artifact. |
| paradox_list | 36 quote crosswalk rows spanning all 35 paradox IDs. |
| flow | 10-stage ATLAS333 intelligence pipeline (INGESTβSEAL). |
| geometry | Full cognitive geometry map (territories Γ geometries Γ depths). |
TDQS
C2.5/5.0
Scored across 6 tools
Disambiguation4/5
The tools have distinct names for different cognitive stages (init, observe, think, judge, memory, route), but without detailed descriptions, the boundary between 'think' and 'judge' could be ambiguous to an agent. Overall, the purposes are mostly distinct.
Naming Consistency5/5
All tools follow a consistent 'arif_' prefix followed by a single verb or noun, creating a predictable and uniform naming pattern throughout.
Tool Count5/5
Six tools is a well-scoped number for a cognitive agent loop, covering essential functions (initialization, perception, reasoning, decision, memory, routing) without redundancy or bloat.
Completeness4/5
The set covers a full cognitive cycle from initialization to routing, but lacks an explicit action execution or learning tool, leaving minor gaps that agents might need to work around.
Maintenance
ActivityNo data
ResponsivenessSlow