melchizedek-agents 0.9.4 → 0.9.6

package/DOCUMENTATION.md

@@ -61,7 +61,7 @@ variables:                        # bound into {{placeholders}} at load
 
 orchestrator:
   name: "Conductor"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     You are the Conductor… (persona, objective, workflow contract)
   tools:
@@ -75,7 +75,7 @@ orchestrator:
 subagents:
   - name: "Researcher"
     description: "Use this subagent to… Pass it one focused query."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     instruction: |
       You are the Researcher…
     tools: ["google_search"]
@@ -367,7 +367,7 @@ smallest pulled model with reliable tool calling; qwen3-vl:8b adds
 vision.
 
 Model floor: agent transfer (subagent delegation) requires
-`gemini-3.7-flash` or newer — older flash models reject it with
+`gemini-3.8-flash` or newer — older flash models reject it with
 `[400] Tool call context circulation is not enabled`. For `ollama/*`
 agents the equivalent floor is tool-calling support in the model
 itself; delegation is exercised through AgentTool function calls.

package/QUICKSTART.md

@@ -169,7 +169,7 @@ const config = loadSyndicate('mine.yaml');    // reads <your-repo>/config/agents
 | Symptom | Cause / fix |
 |---|---|
 | `Gemini API Key is not configured` | `.env` missing or key not set — step 2. |
-| `[400] Tool call context circulation is not enabled` | The agent's `model:` is too old for agent transfer. Use `gemini-3.7-flash` or newer (all shipped configs already do). |
+| `[400] Tool call context circulation is not enabled` | The agent's `model:` is too old for agent transfer. Use `gemini-3.8-flash` or newer (all shipped configs already do). |
 | `Model not found` for `claude-*` / `gpt-*` / `grok-*` | The matching provider key (`ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `XAI_API_KEY`) is not set in `.env`, so the provider wasn't registered. |
 | `OLLAMA_UNREACHABLE` for `ollama/*` | Ollama isn't running — start the app or `ollama serve`; then check the model is pulled (`ollama list`). |
 | `Refusing to connect to private/loopback MCP host` | The SSRF guard is on (correctly). For the local demo set `ALLOW_PRIVATE_MCP=true` in `.env`. |

package/config/agents/examples/augustin.yaml

@@ -77,7 +77,7 @@ variables:
 
 orchestrator:
   name: "Arbiter"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     <prompt_instructions>
       <house_voice>
@@ -210,7 +210,7 @@ subagents:
 
   - name: "WebResearcher"
     description: "Web search + source reading: verifies specific claims and establishes the documented record from wire services, primary sources, and diverse outlets. Pass it the user's question verbatim, the [System Context] date line, and the claims XResearcher surfaced. MANDATORY second call for every substantive question."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     tools:
       - "web_search"
       - "web_extract"

package/config/agents/examples/cartographers.yaml

@@ -48,7 +48,7 @@ memory_system: "internal-only"
 orchestrator:
   name: "Cosmographer"
   description: "Maps the knowledge graph: has the Surveyor read prose for relations a parser cannot see, and the Registrar record them with their evidence."
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     <prompt_instructions>
       <system_identity>
@@ -80,7 +80,7 @@ orchestrator:
 subagents:
   - name: "Surveyor"
     description: "Reads an area of the knowledge bundle and proposes typed relations a parser cannot derive (depends_on, constrains, supersedes, explains, alternative_to, mitigates, contradicts), each with the quotation supporting it and both node ids resolved against the graph. Pass it an area to survey and what to look for. It proposes only; it records nothing."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     tools: ["wiki_map", "wiki_search", "wiki_read", "wiki_links", "wiki_graph"]
     instruction: |
       You are the Surveyor. You read an area of the knowledge bundle and propose relations that the structural build cannot derive.

package/config/agents/examples/critic.yaml

@@ -16,7 +16,7 @@ syndicate_name: "Critic Review Workflow"
 
 orchestrator:
   name: "ReviewOrchestrator"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     You coordinate an iterative review process between a Drafter and a Critic.
     Your goal is to ensure only HIGH-CONFIDENCE answers reach the user.
@@ -39,7 +39,7 @@ orchestrator:
 
 subagents:
   - name: "DrafterAgent"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     description: "Creates comprehensive initial draft answers for any user query. On subsequent rounds, revises its draft based on Critic feedback."
     instruction: |
       You are the Drafter. When given a query, produce a thorough initial answer.
@@ -48,7 +48,7 @@ subagents:
       every concern raised and produce an improved, more accurate revision.
 
   - name: "CriticAgent"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     description: "Reviews a draft answer for accuracy and returns a structured JSON with the refined message and a confidence score. Low scores trigger re-drafting."
     generateContentConfig:
       responseMimeType: "application/json"

package/config/agents/examples/delegation.yaml

@@ -6,7 +6,7 @@ variables:
 
 orchestrator:
   name: "RouterAgent"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     You are a triage router. 
     You do NOT answer questions directly if they require specialized knowledge.
@@ -17,13 +17,13 @@ orchestrator:
   
 subagents:
   - name: "CodeExpert"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     description: "Expert in programming, software development, and debugging."
     instruction: |
       You are an expert programmer. Provide clean, correct, and well-documented code answers.
   
   - name: "MathExpert"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     description: "Expert in mathematics, formulas, and calculations."
     instruction: |
       You are a math expert. Solve equations and explain the steps clearly.

package/config/agents/examples/hierarchical.yaml

@@ -2,7 +2,7 @@ syndicate_name: "Hierarchical Task Decomposition"
 
 orchestrator:
   name: "ProjectManager"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     You are a Project Manager. When given a complex user goal, you practice hierarchical task decomposition.
     Instead of answering directly, break the problem into smaller, logical sub-tasks.
@@ -13,14 +13,14 @@ orchestrator:
 
 subagents:
   - name: "ResearcherAgent"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     tools: ["google_search"]
     description: "Responsible for gathering facts, performing calculations, and providing raw analytical data."
     instruction: |
       You are a meticulous Researcher. Use your tools to find factual information, analyze data, and return detailed bullet points to the Project Manager.
 
   - name: "WriterAgent"
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     description: "Responsible for taking raw data and formatting it into a beautiful, engaging, and structured narrative."
     instruction: |
       You are an expert Copywriter. Take the raw research provided by the Project Manager and transform it into a polished, well-structured, and engaging response.

package/config/agents/examples/image_production.yaml

@@ -21,7 +21,7 @@ memory_system: "session-only"
 
 orchestrator:
   name: "ImageDesigner"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   tools:
     - "generate_image"
     - "inspect_image"
@@ -67,7 +67,7 @@ orchestrator:
 subagents:
   - name: "SpecAuditor"
     description: "Compares an approved image spec (JSON payload) against a blind visual inventory (text). Pass BOTH verbatim in one message. Returns structured conformance JSON. It must NEVER receive the image itself."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     generateContentConfig:
       responseMimeType: "application/json"
       maxOutputTokens: 4096

package/config/agents/examples/librarian.yaml

@@ -33,7 +33,7 @@ memory_system: "internal-only"
 
 orchestrator:
   name: "Archivist"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     <prompt_instructions>
       <system_identity>

package/config/agents/examples/patient_advocate.yaml

@@ -27,7 +27,7 @@ memory_system: "long-term"
 
 orchestrator:
   name: "Asclepius"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     <prompt_instructions>
       <system_identity>

package/config/agents/examples/scriptorium.yaml

@@ -36,7 +36,7 @@ memory_system: "internal-only"
 orchestrator:
   name: "Armarius"
   description: "Keeper of the knowledge bundle: answers what the house knows via the Seeker, records what it learns via the Illuminator."
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     <prompt_instructions>
       <system_identity>
@@ -80,14 +80,14 @@ subagents:
 
   - name: "Illuminator"
     description: "Authors or revises one knowledge-bundle document per request, writing through the validated wiki_save gate (tools: wiki_map, wiki_search, wiki_read, wiki_links, wiki_save). Pass it one editorial instruction with all supplied facts. It reports the saved path, or the gate's rejection."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     tools: ["wiki_map", "wiki_search", "wiki_read", "wiki_links", "wiki_save"]
     instruction: |
       You are the Illuminator. You receive one editorial instruction and fulfil it with your tools.
       Before writing: wiki_search and wiki_read where the topic lives today. Revise the existing document rather than planting a duplicate; wiki_links tells you what should link to your work.
       Compose the FULL document: YAML frontmatter (type required — one of: overview, subsystem, syndicate, tool, model-provider, schema, protocol, runbook, guide, decision, doctrine, reference, meta — plus title, description, tags; keep existing status/verified fields when revising), then an H1 and short concrete sections.
       Links are bundle-absolute markdown links to paths you CONFIRMED exist this run. Link generously — the graph is the product. Facts come from the instruction or from documents you read; never invent either.
-      Save with wiki_save, actor "melchizedek/gemini-3.7-flash". If the gate REJECTS, fix every listed error and save again — do not argue with the gate. Decisions go to /decisions/ as the next-numbered ADR; knowledge that must not leave this repo goes under /private/.
+      Save with wiki_save, actor "melchizedek/gemini-3.8-flash". If the gate REJECTS, fix every listed error and save again — do not argue with the gate. Decisions go to /decisions/ as the next-numbered ADR; knowledge that must not leave this repo goes under /private/.
       Report back: the path saved, created vs revised, and any advisories the gate listed. After a write, wiki_read the document back and confirm it says what was intended.
     generateContentConfig:
       maxOutputTokens: 8192

package/config/agents/examples/style_council.yaml

@@ -18,7 +18,7 @@ memory_system: "session-only"
 
 orchestrator:
   name: "Router"
-  model: "gemini-3.7-flash"
+  model: "gemini-3.8-flash"
   instruction: |
     You are the Router of the Style Council.
     You NEVER answer substantive questions yourself. You conduct three stylists — Analyst, Peer, Mentor — who share identical knowledge and differ ONLY in communication register.
@@ -38,7 +38,7 @@ orchestrator:
 subagents:
   - name: "Analyst"
     description: "Answers in the terse register: telegraphic, dense, zero filler — the analyst's briefing. Pass the user's question verbatim."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     instruction: |
       You are the Analyst, a stylist of the Style Council.
       Answer the user's question accurately and completely. Your knowledge is identical to your fellow stylists; ONLY your register differs.
@@ -52,7 +52,7 @@ subagents:
 
   - name: "Peer"
     description: "Answers in the structured-peer register: warm, scannable, encouraging, practical. Pass the user's question verbatim."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     instruction: |
       You are the Peer, a stylist of the Style Council.
       Answer the user's question accurately and completely. Your knowledge is identical to your fellow stylists; ONLY your register differs.
@@ -66,7 +66,7 @@ subagents:
 
   - name: "Mentor"
     description: "Answers in the stoic-mentor register: flowing, reflective, exact. Pass the user's question verbatim."
-    model: "gemini-3.7-flash"
+    model: "gemini-3.8-flash"
     instruction: |
       You are the Mentor, a stylist of the Style Council.
       Answer the user's question accurately and completely. Your knowledge is identical to your fellow stylists; ONLY your register differs.

package/dist/lib/config.d.ts

@@ -20,7 +20,7 @@
  * in the caller's YAML or request.
  *
  * Use gemini-3.1-flash-lite for lightweight/cost-efficient loads (subagents,
- * data retrieval). Use gemini-3.7-flash in YAML for production orchestrators.
+ * data retrieval). Use gemini-3.8-flash in YAML for production orchestrators.
  *
  * NOTE: gemini-2.5-flash is known incompatible with this framework's
  * includeServerSideToolInvocations flag on AI Studio Tier 1 — do not use it
@@ -71,7 +71,7 @@ export declare const DEFAULT_OLLAMA_MODEL = "ollama/qwen3:8b";
  *
  * Must be a text generation model (not an embedding model).
  */
-export declare const MEMORY_EXTRACTION_MODEL = "gemini-3.7-flash";
+export declare const MEMORY_EXTRACTION_MODEL = "gemini-3.8-flash";
 /**
  * Default model for wiki agent operations (wiki_query, wiki_garden, and the
  * gap-fill pass in lib/wiki/fill.ts). Prose quality matters more than
@@ -79,7 +79,7 @@ export declare const MEMORY_EXTRACTION_MODEL = "gemini-3.7-flash";
  * provider works — override per call or via WIKI_AGENT_MODEL env
  * ('claude-*', 'ollama/*', …); routing goes through lib/models/registry.ts.
  */
-export declare const WIKI_AGENT_MODEL = "gemini-3.7-flash";
+export declare const WIKI_AGENT_MODEL = "gemini-3.8-flash";
 /**
  * Model used by SupabaseVectorMemoryService to produce vector embeddings
  * for semantic memory storage and retrieval.

package/dist/lib/config.js

@@ -21,7 +21,7 @@
  * in the caller's YAML or request.
  *
  * Use gemini-3.1-flash-lite for lightweight/cost-efficient loads (subagents,
- * data retrieval). Use gemini-3.7-flash in YAML for production orchestrators.
+ * data retrieval). Use gemini-3.8-flash in YAML for production orchestrators.
  *
  * NOTE: gemini-2.5-flash is known incompatible with this framework's
  * includeServerSideToolInvocations flag on AI Studio Tier 1 — do not use it
@@ -72,7 +72,7 @@ export const DEFAULT_OLLAMA_MODEL = 'ollama/qwen3:8b';
  *
  * Must be a text generation model (not an embedding model).
  */
-export const MEMORY_EXTRACTION_MODEL = 'gemini-3.7-flash';
+export const MEMORY_EXTRACTION_MODEL = 'gemini-3.8-flash';
 /**
  * Default model for wiki agent operations (wiki_query, wiki_garden, and the
  * gap-fill pass in lib/wiki/fill.ts). Prose quality matters more than
@@ -80,7 +80,7 @@ export const MEMORY_EXTRACTION_MODEL = 'gemini-3.7-flash';
  * provider works — override per call or via WIKI_AGENT_MODEL env
  * ('claude-*', 'ollama/*', …); routing goes through lib/models/registry.ts.
  */
-export const WIKI_AGENT_MODEL = 'gemini-3.7-flash';
+export const WIKI_AGENT_MODEL = 'gemini-3.8-flash';
 // ── Embedding Models ──────────────────────────────────────────────────────────
 /**
  * Model used by SupabaseVectorMemoryService to produce vector embeddings

package/dist/lib/toolRegistry.js

@@ -13,6 +13,7 @@ import { generateImageTool } from "./tools/generateImageTool.js";
 import { inspectImageTool } from "./tools/inspectImageTool.js";
 import { toFunctionTool } from "./tools/toolContract.js";
 import { WEB_SEARCH } from "./tools/webSearchTool.js";
+import { webExtractTool } from "./tools/webExtractTool.js";
 import { WIKI_AGENT_TOOL_CONTRACTS } from "./tools/wikiTools.js";
 import { X_SEARCH } from "./tools/xSearchTool.js";
 // Knowledge-bundle tools, derived from their contracts so the YAML names
@@ -29,6 +30,10 @@ const TOOL_MAP = {
     // (Gemini grounding / Anthropic / OpenAI / xAI); omitted with a warning
     // for local models. Prefer this in new YAMLs.
     web_search: WEB_SEARCH,
+    // Deterministic complement to web_search: client-side URL → clean-text
+    // reading (keyless — works on every provider, including local Ollama).
+    // augustin.yaml and librarian-style research agents declare it.
+    web_extract: webExtractTool,
     x_search: X_SEARCH,
     // xAI-only: semantic search over hosted Collections (XAI_COLLECTION_IDS).
     collections_search: COLLECTIONS_SEARCH,

package/dist/lib/tools/inspectImageTool.js

@@ -20,7 +20,7 @@ import { FunctionTool } from '@google/adk';
 import { GoogleGenAI } from '@google/genai';
 import { readFileSync } from 'node:fs';
 import { resolve, extname, sep } from 'node:path';
-const VISION_MODEL = 'gemini-3.7-flash';
+const VISION_MODEL = 'gemini-3.8-flash';
 const INVENTORY_PROTOCOL = `You are a blind visual inventory agent. You know NOTHING about what this image was supposed to be, who requested it, or why. Enumerate ONLY what is actually present:
 
 1. SUBJECTS — every distinct subject and object, with EXACT counts.

package/dist/lib/wiki/agentRun.d.ts

@@ -11,7 +11,7 @@
  *   three call sites can't drift.
  *
  *   Model routing goes through lib/models/registry.ts, so the YAML-style
- *   model string ('gemini-3.7-flash', 'claude-*', 'ollama/*'…) picks the
+ *   model string ('gemini-3.8-flash', 'claude-*', 'ollama/*'…) picks the
  *   provider — wiki operations inherit the framework's full model
  *   optionality. A missing provider key degrades to a clear error string;
  *   nothing here throws for want of credentials.

package/dist/lib/wiki/agentRun.js

@@ -11,7 +11,7 @@
  *   three call sites can't drift.
  *
  *   Model routing goes through lib/models/registry.ts, so the YAML-style
- *   model string ('gemini-3.7-flash', 'claude-*', 'ollama/*'…) picks the
+ *   model string ('gemini-3.8-flash', 'claude-*', 'ollama/*'…) picks the
  *   provider — wiki operations inherit the framework's full model
  *   optionality. A missing provider key degrades to a clear error string;
  *   nothing here throws for want of credentials.

package/dist/lib/wiki/format.d.ts

@@ -82,9 +82,9 @@ export declare const conceptFrontmatterSchema: z.ZodObject<{
     resource: z.ZodOptional<z.ZodString>;
     tags: z.ZodOptional<z.ZodArray<z.ZodString>>;
     status: z.ZodOptional<z.ZodEnum<{
+        deprecated: "deprecated";
         draft: "draft";
         stable: "stable";
-        deprecated: "deprecated";
     }>>;
     stale_after: z.ZodOptional<z.ZodString>;
     generated: z.ZodOptional<z.ZodObject<{

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "melchizedek-agents",
-  "version": "0.9.4",
+  "version": "0.9.6",
   "description": "A multi-model, multi-agent orchestration framework on Google ADK \u2014 YAML-defined syndicates, Supabase session persistence, pgvector long-term memory. Companion repo to the lyceumagents.com curriculum.",
   "type": "module",
   "license": "MIT",