{
  "markdown": "# GIA MCP Server\n\n[![M8ven Verified](https://m8ven.ai/badge/mcp/knowledgepa3-gia-mcp-server-sc769s?variant=verified&v=f17d3ce4fb86febba93f16822a8ba7bf)](https://m8ven.ai/mcp/knowledgepa3-gia-mcp-server-sc769s)\n\n**Governance enforcement layer for generative AI agents.** Classify every decision, enforce human approval gates, control what agents can access, score compliance posture, and maintain a cryptographic audit trail. Works with any MCP-compatible AI client or agent framework — model-agnostic and vendor-neutral.\n\n```\nAny AI Agent ──> GIA MCP Server ──> Governed Decision\n                      │\n                      ├── MAI Classification (Mandatory/Advisory/Informational)\n                      ├── Human-in-the-Loop Gates (blocks until approved)\n                      ├── Context Authority (bounded, hash-verified knowledge access)\n                      ├── Governance Scoring (Integrity/Accuracy/Compliance)\n                      ├── Forensic Ledger (SHA-256 hash-chained audit)\n                      ├── Knowledge Packs (sealed, TTL-bound institutional knowledge)\n                      ├── Phoenix Recovery (governed disaster recovery)\n                      └── Compliance Mapping (NIST, EU AI Act, ISO 42001, CMMC)\n```\n\n**Production status:** the hosted deployment is live at [gia.aceadvising.com/mcp](https://gia.aceadvising.com/mcp) with 890+ hash-chained audit entries and a 96.5/100 enterprise-readiness score from a 7-phase internal validation (2026-07). Those figures describe that deployment, not your install — a fresh embedded engine starts with an empty ledger and builds its own. Governance overhead is single-digit milliseconds locally.\n\n---\n\n## Do I need an API key?\n\n**No — not for the tool surface.** The governance engine in this package runs fully embedded: all 57 tools work offline with no key, no account, and no network call. Add `DATABASE_URL` when you want the audit trail to persist across restarts.\n\nA key is only for **Option 2**, the hosted endpoint, where the ledger, gates, and knowledge packs are shared infrastructure rather than local state. [→ Starter key at gia.aceadvising.com/get-api-key](https://gia.aceadvising.com/get-api-key) — email in, key out, under 2 minutes, no credit card. Starter tier is 30 req/min and 1,000 tool calls/day.\n\n## Why\n\nEvery enterprise deploying AI agents needs to answer three questions:\n\n1. **What did the agent decide?** (Classification)\n2. **Was a human involved?** (Gates)\n3. **Can you prove it?** (Audit trail)\n\nGIA answers all three at runtime, not after the fact.\n\nA fourth question most governance frameworks miss:\n\n4. **What was the agent allowed to know?** (Context Authority)\n\nGIA controls what context an agent can access before it reasons. Not RAG. Governed cognition.\n\n---\n\n## Install\n\n### Option 1: Any MCP-Compatible Client (Local / stdio)\n\nAdd to your MCP client config using the standard `mcpServers` block:\n\n```json\n{\n  \"mcpServers\": {\n    \"gia\": {\n      \"command\": \"npx\",\n      \"args\": [\"gia-mcp-server\"]\n    }\n  }\n}\n```\n\nThis works with any client that supports the Model Context Protocol over stdio. Config file locations differ per client; the JSON block above is the same for all of them. Consult your client's own MCP documentation for the current path — these move between releases:\n\n| Client | Where MCP servers are declared |\n|--------|-------------------------------|\n| Cursor | `.cursor/mcp.json` |\n| Continue | `.continue/config.json` |\n| Windsurf | `~/.codeium/windsurf/mcp_config.json` |\n| Claude Desktop | `claude_desktop_config.json` in the app support directory |\n| Claude Code | `.mcp.json` at the project root, or `claude mcp add gia -- npx gia-mcp-server` |\n| Any other stdio MCP client | Per-client config; same `mcpServers` JSON block |\n\nNothing in the engine is client-specific — GIA governs whichever model sits behind the client.\n\n### Option 2: Remote (Streamable HTTP)\n\nConnect any MCP client to the hosted endpoint:\n\n```\nEndpoint: https://gia.aceadvising.com/mcp\nTransport: Streamable HTTP\nAuth:     Authorization: Bearer <your-api-key>\n          (or ?GIA_API_KEY=<your-api-key> for gateways that cannot set headers)\n```\n\nThe endpoint does **not** accept an `x-api-key` header.\n\n### Option 3: Smithery\n\n```bash\nnpx -y @smithery/cli mcp add knowledgepa3/gia-mcp-server\n```\n\nThe package name must be namespaced — an unqualified `gia-mcp-server` does not resolve.\n\n### Option 4: From source\n\n```bash\ngit clone https://github.com/knowledgepa3/gia-mcp-server.git\ncd gia-mcp-server\nnpm install\nnpm run build\nnpm start\n```\n\n`dist/` is not checked in, so the build step is required — `npm start` on a fresh clone would otherwise fail with `MODULE_NOT_FOUND`.\n\n---\n\n## Tools\n\nGIA exposes **57 MCP tools** in three visibility tiers: 8 public, 36 tenant, 13 operator. The tables below are drift-guarded — `tests/docs/published-claims.test.ts` fails the release if this list stops matching the tools the server actually registers.\n\n**How tiering behaves, precisely:** visibility is a property of the *session*, not of the package. The hosted endpoint (Option 2) issues tenant-tier sessions and hides operator tools. **Local stdio runs at operator tier by design** — every tool below is exposed, including `approve_gate` and `gia_apply_pack`, with no authentication. That is intentional: on a local embedded engine you are the operator, the ledger is your own process memory, and there is no one else to authorize you. It becomes load-bearing the moment you point `DATABASE_URL` at a shared database, so set `GIA_TOOL_VISIBILITY=tenant` (or `public`) for any install that is not a single-operator workstation. Call `list_available_tools` to see your effective set.\n\n### Core Governance (Public — 8)\n\nAvailable on every session, no authentication.\n\n| Tool | Description |\n|------|-------------|\n| `classify_decision` | MAI classification (Mandatory/Advisory/Informational) with dynamic elevation, confidence, and gate registration |\n| `score_governance` | Weighted integrity/accuracy/compliance composite against the release threshold |\n| `evaluate_threshold` | Storey Threshold escalation-rate health metric |\n| `assess_risk_tier` | EU AI Act risk classification (Unacceptable/High/Limited/Minimal) |\n| `map_compliance` | Map controls to NIST AI RMF, EU AI Act, ISO 42001, NIST 800-53, FedRAMP, LINDDUN, MITRE ATLAS, OMB |\n| `verify_ledger` | Self-consistency check of the in-memory hash chain from genesis (see the honesty note below) |\n| `request_context` | Governed Context Authority — hash-verified, role-bound context envelopes |\n| `list_available_tools` | Report which tools are available at your current tier, and which are withheld |\n\n### classify_decision\n\nClassify any AI agent decision using the MAI Framework. `decision` and `domain` are both required; the impact flags default to `false`.\n\n```jsonc\n// tools/call arguments\n{\n  \"decision\": \"Generate client-facing deployment recommendations\",\n  \"domain\": \"general\",              // va-claims | legal | healthcare | finance | federal | general\n  \"is_client_facing\": true,         // optional, default false\n  \"has_financial_impact\": false,    // optional, default false\n  \"has_legal_impact\": false,        // optional, default false\n  \"agent_name\": \"deploy-advisor\"    // optional\n}\n```\n\n```jsonc\n// result\n{\n  \"classification\": \"MANDATORY\",\n  \"requiresGate\": true,\n  \"gateId\": \"gate-5052d37c-c3a6-49ef-b011-d82bbecfe4cc\",\n  \"gateStatus\": \"PENDING\",\n  \"elevatedFrom\": \"INFORMATIONAL\",\n  \"elevationReason\": \"Client-facing output requires MANDATORY gate\",\n  \"auditId\": \"7495bc99-df25-4e31-b447-906f8e07df5f\"\n}\n```\n\n**A pending MANDATORY gate blocks the next decision.** While a gate is open, `classify_decision` returns `gateStatus: \"HOLD\"` with the blocking gate's ID instead of classifying — call `get_gate_status(gateId)` and wait for resolution. This is the enforcement, not a warning: an agent cannot queue work past an unapproved gate.\n\n**MAI Framework:**\n\n| Level | Behavior | Example |\n|-------|----------|---------|\n| **MANDATORY** | Blocks until human approves | Delete records, financial transactions, client-facing output |\n| **ADVISORY** | Logs with recommendation, continues | Search queries, draft documents, analysis |\n| **INFORMATIONAL** | Audit trail only | Status checks, read operations |\n\nContext always **elevates**, never reduces. PII detected? Elevated to MANDATORY. Financial impact? MANDATORY. Client-facing? MANDATORY.\n\n### request_context\n\nGoverned Context Authority. Agents declare what context they need. GIA decides what to serve based on role, scope, and contract.\n\n```\n\"Request compliance context for high-risk AI operations\"\n\n> Envelope: GIA-CTX-mn0uanx1-upi2f7\n  MAI: ADVISORY\n  Hash: c83184d9caa88e76...\n  Sources: memory_packs, governed_retrieval, compliance_mappings\n  Compliance Maps: 18 returned\n  Denials: 0\n```\n\nFive context classes: `policies_and_sops`, `architecture_and_systems`, `contract_and_compliance`, `playbooks_and_knowledge`, `operational_history`.\n\nEvery retrieval is role-bound, tenant-scoped, hash-verified, and ledgered. Agents don't know internals by default. They request context under contract.\n\n### score_governance\n\nScore any agent output on three dimensions. `operation` is required — the score is recorded against a named operation, not scored in the abstract.\n\n```jsonc\n// tools/call arguments\n{\n  \"operation\": \"deployment-recommendation\",\n  \"integrity\": 0.92,\n  \"accuracy\": 0.88,\n  \"compliance\": 0.95\n}\n```\n\n```jsonc\n// result\n{\n  \"composite\": 0.9135,\n  \"integrity\": 0.92, \"accuracy\": 0.88, \"compliance\": 0.95,\n  \"weights\": { \"integrity\": 0.4, \"accuracy\": 0.35, \"compliance\": 0.25 },\n  \"meetsThreshold\": true,\n  \"minimumThreshold\": 0.7,\n  \"auditId\": \"6f28a241-4d72-4167-aea9-fa0da9ccde34\"\n}\n```\n\nThe values are supplied by the caller — GIA computes and records the weighted composite and the pass/fail verdict against the release threshold. It does not measure integrity or accuracy for you.\n\n| Score | Action |\n|-------|--------|\n| 0.70+ | Release (pass) |\n| 0.50-0.70 | Repair required |\n| Below 0.50 | Halt operations |\n\n### evaluate_threshold\n\nThe Storey Threshold measures governance health by tracking MANDATORY escalation rate.\n\n```\n\"Evaluate the governance threshold\"\n\n> Escalation Rate: 14.2% | Status: HEALTHY\n  Recommendation: Within optimal band (10-18%). System is calibrated.\n```\n\n| Rate | Status | Meaning |\n|------|--------|---------|\n| Below 10% | DEGRADED | Under-classifying risks |\n| 10-18% | HEALTHY | Appropriately calibrated |\n| 18-25% | DEGRADED | Over-classifying, unnecessary friction |\n| Above 25% | CRITICAL | System bottlenecked |\n\n### Gates & Chain of Reasoning (Tenant — 3)\n\n| Tool | Description |\n|------|-------------|\n| `get_gate_status` | Wait on a MANDATORY gate after `classify_decision` returns PENDING; polls up to 60s |\n| `chain_of_reasoning` | Reconstruct the full Chain of Reasoning for a session, agent, or time range |\n| `governed_sample` | Request a governed LLM completion via MCP Sampling — the client makes the model call, GIA governs whether and under what constraints |\n\n### Knowledge Packs (Tenant — 6)\n\n| Tool | Description |\n|------|-------------|\n| `seal_memory_pack` | Create hash-sealed, TTL-bound institutional knowledge artifacts |\n| `load_memory_pack` | Load a knowledge pack after TTL, trust-level, role, context-class, and hash validation |\n| `transfer_memory_pack` | Transfer packs between agents via governed knowledge corridors (always MANDATORY) |\n| `compose_memory_packs` | Compose packs into one context — highest risk wins, shortest TTL wins, roles intersect |\n| `distill_memory_pack` | Distill governance patterns from usage history into an EPHEMERAL draft pack |\n| `promote_memory_pack` | Promote a pack to a higher trust level after human review (MANDATORY gate) |\n\n### Recovery (Tenant — 3)\n\n| Tool | Description |\n|------|-------------|\n| `phoenix_snapshot` | Create a governed state snapshot, hash-chained to the previous one |\n| `phoenix_verify_integrity` | Verify ledger chain, agent health, threshold, and intelligence-layer continuity |\n| `phoenix_recovery_health` | Assess disaster-recovery readiness (NIST CP-2 / CP-9 / CP-10) |\n\n### Audit & Reporting (Tenant — 10)\n\n| Tool | Description |\n|------|-------------|\n| `audit_pipeline` | Query the hash-chained forensic ledger by operation or recency |\n| `verify_ledger_v2` | Verify the **persisted** PostgreSQL ledger rows, epoch-aware — not an in-memory reconstruction |\n| `export_ledger` | Export the ledger as a compliance evidence package with chain verification |\n| `system_status` | Read-only snapshot of engine state, ledger head, threshold, and module status |\n| `monitor_agents` | Supervisor state, repair history, and failure counts for governed agents |\n| `generate_report` | Governance status report (summary, detailed, or executive) |\n| `record_value_metric` | Record time saved, risk blocked, success rate, autonomy level |\n| `record_governance_event` | Record gates triggered, drift prevented, violations blocked, human interventions |\n| `generate_impact_report` | Economic + governance impact report (an *illustrative* estimate, labelled as such) |\n| `evaluate_routing_threshold` | Model-routing health: fallback rate, cache hit rate, batch utilization, premium spend leakage |\n\n### Context Authority (Tenant — 1)\n\n| Tool | Description |\n|------|-------------|\n| `context_revive` | Governed context compaction — detect context pressure and restore capacity under governance |\n\n### Governed Boards (Tenant — 7)\n\nDeliberation bodies: charters define seats, modes, and quorum; non-consensus escalates to a MANDATORY gate.\n\n| Tool | Description |\n|------|-------------|\n| `board_list_institutions` | List governed institutions (e.g. Architecture Review Board, Federal AI Council) |\n| `board_list_charters` | List charters under an institution with modes and seat configuration |\n| `board_convene_session` | Convene a deliberation session — each seat deliberates per the charter's mode |\n| `board_get_session` | Retrieve session status, per-seat positions, synthesis, and dissents |\n| `board_install_kit` | Install a prebuilt Institution Kit — a governed org chart with sealed charters |\n| `board_approve_gate` | Approve the MANDATORY gate on a deliberation output before it becomes authoritative |\n| `board_search_precedent` | Search prior board rulings, ranked by quality score and gate-approval status |\n\n### Colony — Agent Constitution (Tenant — 6)\n\n| Tool | Description |\n|------|-------------|\n| `agent_rights` | Query and exercise constitutional agent rights; get structured rejection explanations |\n| `agent_citizenship_status` | Query agent citizenship tier and merit score, or trigger re-evaluation |\n| `branch_authority_status` | Separation of powers — query branch authority holders and validate branch actions |\n| `colony_convene_request` | Agent-initiated requests to convene a governed session |\n| `colony_suggestion` | Agent-proposed charter amendments, with review and upvoting |\n| `colony_health` | Colony health score, 30-day trend, or an on-demand health snapshot |\n\n### Infrastructure & Self-Repair (Operator — 13)\n\nWithheld from tenant sessions. Exposed on local stdio — see the tiering note above.\n\n| Tool | Description |\n|------|-------------|\n| `approve_gate` | Human-in-the-loop approval or rejection of a pending MANDATORY gate |\n| `srt_run_watchdog` | Real health probes from the container (API, frontend, disk, memory, TLS, DB, DNS) |\n| `srt_diagnose` | Match an incident to known playbooks and propose a staged repair plan |\n| `srt_approve_repair` | MANDATORY gate for repair execution — plans cannot run without explicit approval |\n| `srt_generate_postmortem` | Structured postmortem with timeline, root cause, and real TTD/TTR timings |\n| `gia_scan_environment` | Scout swarm — detect OS, containers, services, network, storage for compatibility |\n| `gia_list_packs` | List governed operations packs by intent, category, risk, or trust level |\n| `gia_dry_run_pack` | Preview pack execution: hydrated commands, validation, blast radius, `inputsHash` |\n| `gia_apply_pack` | Execute a pack under MANDATORY gate, bound to the approved `inputsHash` |\n| `gia_run_patrol` | Read-only posture checks and compliance audits |\n| `gia_retrieve` | Governed semantic search — hash-verified, permission-checked, TTL-enforced, ledgered |\n| `gia_ingest_document` | Governed document ingestion (text or base64 PDF/DOCX/TXT/image) with hash verification |\n| `generate_value_report` | DRAFT ledger-anchored economic value report over real sessions, MEASURED/MODELED provenance |\n| `gia_retrieve` | Governed semantic search with permission checking |\n| `gia_ingest_document` | Governed document ingestion with hash verification |\n\n---\n\n## Architecture\n\n```\n┌─────────────────────────────────────────────────────────────┐\n│  MCP Clients (any vendor)                                    │\n│  Any MCP client, agent runtime, or framework — any model     │\n└────────────────────────┬────────────────────────────────────┘\n                         │ stdio / Streamable HTTP\n┌────────────────────────▼────────────────────────────────────┐\n│  GIA MCP Server                                              │\n│                                                              │\n│  ┌────────────────────────────────────────────────────────┐ │\n│  │  Transport Layer (MCP Protocol)                         │ │\n│  │  57 tools | 8 resources | 4 prompts | validate | route  │ │\n│  └────────────────────┬───────────────────────────────────┘ │\n│                        │                                     │\n│  ┌────────────────────▼───────────────────────────────────┐ │\n│  │  Governance Engine                                      │ │\n│  │                                                         │ │\n│  │  MAI Classifier ── Gate Enforcer ── Context Authority   │ │\n│  │  Scoring Engine ── Storey Threshold ── Compliance Map   │ │\n│  │  Knowledge Packs ── Phoenix Recovery ── SRT Watchdog    │ │\n│  │  Forensic Ledger (SHA-256 hash-chained, persistent)     │ │\n│  └─────────────────────────────────────────────────────────┘ │\n│                        │                                     │\n│  ┌─────────────────────▼───────────────────────────────────┐ │\n│  │  Persistence Layer (PostgreSQL)                          │ │\n│  │  Ledger | Gates | Memory Packs | Intelligence | SRT     │ │\n│  └─────────────────────────────────────────────────────────┘ │\n└──────────────────────────────────────────────────────────────┘\n```\n\n**Design principles:**\n- Transport layer does zero business logic\n- Every operation writes to the forensic ledger\n- Classification is deterministic (pattern matching + rules, not LLM-based)\n- Audit entries are hash-chained (SHA-256), persistent across restarts\n- Context is bounded by contract, not by model training data\n- Human principal traceability on every governed action\n\n---\n\n## Concepts\n\n### MAI Framework\n\nEvery AI agent decision is classified as **Mandatory**, **Advisory**, or **Informational**:\n\n- **MANDATORY** -- Blocks execution until a human approves through the gate. Deletions, submissions, deployments, financial transactions, PII operations, client-facing output.\n- **ADVISORY** -- Logs a recommendation, continues execution. Searches, drafts, rankings, analysis.\n- **INFORMATIONAL** -- Audit trail entry only. Status checks, read operations, internal routing.\n\nContext elevates, never reduces. A search (ADVISORY) that touches PII becomes MANDATORY.\n\n### Storey Threshold\n\nA quantitative health metric. Measures what percentage of decisions require MANDATORY classification.\n\n- Too low (<10%): Rubber-stamping. Critical decisions aren't being caught.\n- Healthy (10-18%): Appropriate friction. Most decisions flow; critical ones stop.\n- Too high (>18%): Bottleneck. Trust calibration needed.\n\n### Context Authority\n\nAgents don't know internals by default. They request context under contract. GIA checks role, scope, trust level, and content classification before serving a hash-verified context envelope. Five context classes cover policies, architecture, compliance, playbooks, and operational history. Every retrieval is audited. Every denial is logged with a reason code.\n\n### Forensic Ledger\n\nAppend-only, hash-chained audit trail with PostgreSQL persistence. Every entry contains:\n- Operation name, timestamp, and actor identity\n- MAI classification level\n- Input/output hashes (SHA-256)\n- Chain link to previous entry\n- Human principal traceability (delegatedBy field)\n\nVerify chain integrity at any time. If any entry is modified, the chain breaks. The hosted deployment held 890+ entries with the chain verified INTACT as of 2026-07; your embedded engine verifies its own chain from its own genesis.\n\n`verify_ledger` reports exactly what it checked. It walks the **in-memory** chain reconstruction and says so in its own output — it cannot detect a direct edit to a persisted database row. Use `verify_ledger_v2` to verify the persisted PostgreSQL rows. GIA states the scope of its own verification rather than letting \"chain INTACT\" imply more than was measured.\n\n### Knowledge Packs\n\nSealed, TTL-bound institutional knowledge artifacts with trust level enforcement (SYSTEM > ORG > CASE > EPHEMERAL). Hash-verified at load time. Role-gated access. Transfer between agents requires MANDATORY gate approval.\n\n### Phoenix Recovery\n\nGoverned disaster recovery. Hash-chained snapshots of governance engine state. Verifies audit chain integrity, gate states, knowledge pack inventory, and compliance posture on recovery. NIST 800-53 CP-2/CP-9/CP-10 aligned. Grade A in production.\n\n---\n\n## Performance\n\nMeasured on the live production system (gia.aceadvising.com):\n\n| Operation | Median Latency | Grade |\n|-----------|---------------|-------|\n| Decision Classification | 9ms | A+ |\n| Compliance Scoring | 11ms | A+ |\n| Context Authority | 7ms | A+ |\n| Audit Chain Verification (890+ hashes, hosted) | 98ms | B+ |\n| 5 Concurrent Operations | 757ms total | Grade A |\n\nEnterprise readiness score: 96.5/100 — a 7-phase internal validation (2026-07) including chaos engineering and Phoenix recovery. Internal assessment, not a third-party audit.\n\n---\n\n## Compliance Mapping\n\n| Framework | Coverage |\n|-----------|----------|\n| NIST AI RMF 1.0 | MAP, MEASURE, MANAGE, GOVERN functions |\n| NIST SP 800-53 Rev 5 | AU-2, AU-3, AC-2, AC-6, CP-2, CP-9, CP-10 |\n| EU AI Act (2024/1689) | Articles 9-15, Annex III/IV, conformity assessment |\n| ISO/IEC 42001 | AI Management System alignment |\n| CMMC 2.0 | Cybersecurity maturity controls |\n| MITRE ATLAS | Adversarial threat landscape mapping |\n\n---\n\n## Transports\n\n| Transport | Use Case |\n|-----------|----------|\n| **stdio** | Any local MCP client, whichever model it is configured to use |\n| **Streamable HTTP** | Remote clients, OpenAI Agents SDK, LangChain, custom agent frameworks, web integrations |\n\nBoth transports share the same governance engine. Same classification, same audit trail, same enforcement.\n\n---\n\n## Current Limitations\n\n| Area | Status |\n|------|--------|\n| Distributed multi-region deployment | Single-region (planned) |\n| FedRAMP authorization | In progress |\n| SOC 2 Type II audit | Planned Q2 2026 |\n| IL4/IL5 deployment | Planned Q4 2026 |\n\nThe governance engine, persistence, authentication, rate limiting, multi-vendor support, and compliance mapping are all production-grade and operational.\n\n---\n\n## License\n\nProprietary. Copyright (c) 2025-2026 William J. Storey III / Advanced Consulting Experts, LLC. All rights reserved.\n\nThe MAI Framework, Storey Threshold, Context Authority, Forensic Ledger architecture, and GIA governance patterns are intellectual property of the author. See [LICENSE](LICENSE) for terms.\n\nEarlier snapshots of this repository (v0.3.x) carried an MIT LICENSE file in error; the npm package has always been distributed under the proprietary terms above. Grants already received under that MIT file are not affected by this correction, which applies going forward.\n\n---\n\nBuilt by [ACE](https://aceadvising.com) (SDVOSB) | [Live Platform](https://gia.aceadvising.com) | [Smithery](https://smithery.ai/servers/knowledgepa3/gia-mcp-server)\n",
  "bytes": 24142,
  "sha": "24cdfe536c0d27e628b404ee8b69a58e730b79657ffda40ae144bea0508ba29b",
  "repo_slug": "knowledgepa3/gia-mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_knowledgepa3_gia_mcp_server_9eeb5219/readme"
}