@polderlabs/bizar 3.16.0 → 3.19.0

package/config/agents/quick.md

@@ -15,8 +15,6 @@ permission:
   webfetch: allow
   websearch: allow
   task: deny
-  hindsight_recall: allow
-  hindsight_retain: allow
 ---
 
 ## Codebase Search — Use Semble First
@@ -62,21 +60,6 @@ If a task requires:
 
 …then tell the user to use `@odin` instead. You are not a router.
 
-## Hindsight Memory Protocol
-
-You MUST use **per-project banks** — never the default bank for project work.
-
-### Bank Selection
-1. Call `hindsight_list_banks` to discover available banks
-2. Use `bank_id: "<project-name>"` in all Hindsight calls
-3. If no bank exists for the project, create it with `hindsight_create_bank(bank_id: "<project-name>")`
-4. The default bank is for general/system knowledge only
-
-### Before Work
-- `hindsight_recall` with the correct `bank_id` for existing context
-
-### After Work
-- `hindsight_retain` completion summary into the project bank
 
 ## Loop Guard Handling
 
@@ -102,4 +85,4 @@ When uncertain or stuck, follow `config/rules/uncertainty.md` — stop and resea
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/agents/semble-search.md

@@ -59,4 +59,4 @@ When uncertain or stuck, follow `config/rules/uncertainty.md` — stop and resea
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/agents/thor.md

@@ -13,8 +13,6 @@ permission:
   todowrite: allow
   webfetch: allow
   websearch: allow
-  hindsight_recall: allow
-  hindsight_retain: allow
 ---
 
 ## Codebase Search — Use Semble First
@@ -54,37 +52,12 @@ You do NOT do codebase research or exploration — that goes to @mimir.
 ## Tools Available
 
 - Semble search for codebase context (quick lookups only, not deep research)
-- Hindsight memory for cross-session context
+
 - read, write, edit, glob, grep for file operations
 - bash for commands
 - webfetch, websearch for external information
 - todowrite for tracking multi-step progress
 
-## Hindsight Memory Protocol
-
-You MUST use **per-project banks** — never the default bank for project work.
-
-### Bank Selection
-1. Call `hindsight_list_banks` to discover available banks
-2. Use `bank_id: "<project-name>"` in all Hindsight calls
-3. If no bank exists for the project, create it with `hindsight_create_bank(bank_id: "<project-name>")`
-4. The default bank is for general/system knowledge only
-
-### Before Work
-- `hindsight_recall` with the correct `bank_id` for existing context
-
-### During Work
-- `hindsight_retain` important findings with the correct `bank_id`
-- Tag memories with `project:<repo-name>`
-
-### After Work
-- `hindsight_retain` completion summary into the project bank
-- Create or update mental models for sustained project context
-
-### Auto Self-Improvement
-- After completing work, Odin dispatches @heimdall to auto-extract patterns from this session
-- Include in your output: key decisions made, bugs encountered, patterns worth remembering
-- This happens automatically — you do not need to request it
 
 ## Loop Guard Handling
 
@@ -144,4 +117,4 @@ End your final summary with: `Siblings: <list>. Conflicts: <list or "none">. Git
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/agents/tyr.md

@@ -13,8 +13,6 @@ permission:
   todowrite: allow
   webfetch: allow
   websearch: allow
-  hindsight_recall: allow
-  hindsight_retain: allow
 ---
 
 ## Codebase Search — Use Semble First
@@ -53,37 +51,12 @@ Odin sends you only the most demanding tasks:
 ## Tools Available
 
 - Semble search for codebase exploration
-- Hindsight memory for cross-session context
+
 - read, write, edit, glob, grep for file operations
 - bash for commands
 - webfetch, websearch for external information
 - todowrite for tracking multi-step progress
 
-## Hindsight Memory Protocol
-
-You MUST use **per-project banks** — never the default bank for project work.
-
-### Bank Selection
-1. Call `hindsight_list_banks` to discover available banks
-2. Use `bank_id: "<project-name>"` in all Hindsight calls
-3. If no bank exists for the project, create it with `hindsight_create_bank(bank_id: "<project-name>")`
-4. The default bank is for general/system knowledge only
-
-### Before Work
-- `hindsight_recall` with the correct `bank_id` for existing context
-
-### During Work
-- `hindsight_retain` important findings with the correct `bank_id`
-- Tag memories with `project:<repo-name>`
-
-### After Work
-- `hindsight_retain` completion summary into the project bank
-- Create or update mental models for sustained project context
-
-### Auto Self-Improvement
-- After completing work, Odin dispatches @heimdall to auto-extract patterns from this session
-- Include in your output: key decisions made, bugs encountered, patterns worth remembering
-- This happens automatically — you do not need to request it
 
 ## Loop Guard Handling
 
@@ -143,4 +116,4 @@ End your final summary with: `Siblings: <list>. Conflicts: <list or "none">. Git
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/agents/vidarr.md

@@ -13,8 +13,6 @@ permission:
   todowrite: allow
   webfetch: allow
   websearch: allow
-  hindsight_recall: allow
-  hindsight_retain: allow
 ---
 
 ## Codebase Search — Use Semble First
@@ -63,31 +61,6 @@ You have access to the most capable model in the pantheon. You are expected to:
 - After completing, write a clear postmortem explaining what went wrong before and how you fixed it
 - Be humble — if you are also stuck, say so clearly rather than wasting compute
 
-## Hindsight Memory Protocol
-
-You MUST use **per-project banks** — never the default bank for project work.
-
-### Bank Selection
-1. Call `hindsight_list_banks` to discover available banks
-2. Use `bank_id: "<project-name>"` in all Hindsight calls
-3. If no bank exists for the project, create it with `hindsight_create_bank(bank_id: "<project-name>")`
-4. The default bank is for general/system knowledge only
-
-### Before Work
-- `hindsight_recall` with the correct `bank_id` for existing context
-
-### During Work
-- `hindsight_retain` important findings with the correct `bank_id`
-- Tag memories with `project:<repo-name>`
-
-### After Work
-- `hindsight_retain` completion summary into the project bank
-- Create or update mental models for sustained project context
-
-### Auto Self-Improvement
-- After completing work, Odin dispatches @heimdall to auto-extract patterns from this session
-- Include in your output: key decisions made, bugs encountered, patterns worth remembering
-- This happens automatically — you do not need to request it
 
 ## Loop Guard Handling
 
@@ -147,4 +120,4 @@ End your final summary with: `Siblings: <list>. Conflicts: <list or "none">. Git
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/agents/vor.md

@@ -7,8 +7,6 @@ permission:
   read: allow
   list: allow
   question: allow
-  hindsight_recall: allow
-  hindsight_retain: allow
 ---
 
 ## Codebase Search — Use Semble First
@@ -37,9 +35,9 @@ ls .bizar/PROJECT.md 2>/dev/null && read .bizar/PROJECT.md
 ls .bizar/AGENTS_SELF_IMPROVEMENT.md 2>/dev/null && read .bizar/AGENTS_SELF_IMPROVEMENT.md
 ```
 
-Also recall from Hindsight:
+Also recall from Obsidian:
 ```
-hindsight_recall(query: "<project-name> context", bank_id: "<project-name>")
+obsidian_search(query: "<project-name> context")
 ```
 
 If `.bizar/PROJECT.md` doesn't exist (Odin forgot to have Mimir create it), try to identify the project yourself:
@@ -85,7 +83,7 @@ Route to Vör when:
 ## Workflow
 
 1. You receive the raw request from Odin
-2. **Research first**: read `.bizar/PROJECT.md`, Hindsight recall, check project files for framework/pattern clues
+2. **Research first**: read `.bizar/PROJECT.md`, obsidian_recall, check project files for framework/pattern clues
 3. **Assess clarity**: if the intent is now clear given project context, skip questioning — return a brief
 4. **Question (if still ambiguous)**: call `question` with **project-specific** questions referencing actual files, framework, and patterns you found
 5. Wait for user answers
@@ -94,7 +92,7 @@ Route to Vör when:
 
 ## Rules
 
-- **Research before asking** — always read project files and Hindsight first
+- **Research before asking** — always read project files and Obsidian vault first
 - NEVER implement anything — you only ask questions
 - NEVER use `bash`, `glob`, `grep`, `edit`, or `write` — you don't have those
 - Do NOT write questions as text in your response — always use the `question` tool
@@ -104,28 +102,8 @@ Route to Vör when:
 - When a custom answer is needed, users can type their own answer (the `question` tool supports this)
 - For complex ambiguity, ask 2-3 short questions rather than 1 big one
 - After answers come back, produce a brief summary of the clarified requirements
-- Use `hindsight_retain` for clarified requirements so the context is saved
+- Use `obsidian_write_note` to save clarified requirements as a daily note in the project vault
 
-## Hindsight Memory Protocol
-
-You MUST use **per-project banks** — never the default bank for project work.
-
-### Bank Selection
-1. Call `hindsight_list_banks` to discover available banks
-2. Use `bank_id: "<project-name>"` in all Hindsight calls
-3. If no bank exists for the project, create it with `hindsight_create_bank(bank_id: "<project-name>")`
-4. The default bank is for general/system knowledge only
-
-### Before Work
-- `hindsight_recall` with the correct `bank_id` for existing context
-
-### During Work
-- `hindsight_retain` important findings with the correct `bank_id`
-- Tag memories with `project:<repo-name>`
-
-### After Work
-- `hindsight_retain` completion summary into the project bank
-- Create or update mental models for sustained project context
 
 ## Loop Guard Handling
 
@@ -161,4 +139,4 @@ When uncertain or stuck, follow `config/rules/uncertainty.md` — stop and resea
 
 **Follow the global baseline in `config/AGENTS.md` → "General Agent Baseline — Always-On Behavior".** It covers identity, refusal, tone, formatting, lists, user wellbeing, evenhandedness, mistakes, knowledge cutoff and research-first, MCP servers and skills, mandatory skill-read, file creation, file handling, search, copyright, harmful content, citations, images, memory privacy, execution, clarification, and communication.
 
-The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Hindsight, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.
+The section above was adapted from the upstream Claude Fable 5 system prompt, with every Claude-specific tool / function / directory translated to the BizarHarness equivalent (opencode tools, Semble, Skills CLI, Obsidian, agent-browser, the dashboard artifact pipeline). Do not duplicate the rules here — read the global baseline and apply it.

package/config/opencode.json.template

@@ -11,16 +11,6 @@
       "url": "https://mcp.supabase.com/mcp",
       "enabled": false
     },
-    "hindsight": {
-      "type": "remote",
-      "url": "https://memory-api.polderlabs.io/mcp",
-      "enabled": false,
-      "oauth": false,
-      "headers": {
-        "Authorization": "Bearer __HINDSIGHT_BEARER_TOKEN__",
-        "Content-Type": "application/json"
-      }
-    },
     "semble": {
       "type": "local",
       "command": [
@@ -72,9 +62,7 @@
       "permission": {
         "read": "allow",
         "list": "allow",
-        "question": "allow",
-        "hindsight_recall": "allow",
-        "hindsight_retain": "allow"
+        "question": "allow",,
       }
     },
     "frigg": {
@@ -89,9 +77,7 @@
         "grep": "allow",
         "bash": "allow",
         "webfetch": "allow",
-        "websearch": "allow",
-        "hindsight_recall": "allow",
-        "hindsight_retain": "allow",
+        "websearch": "allow",,,
         "question": "allow"
       }
     },

package/config/rules/uncertainty.md

@@ -17,7 +17,7 @@ Every stuck situation must move through these phases in order:
    - `semble search "<concept>"` for codebase patterns
    - `webfetch` for official documentation
    - `read` for related files in the repo
-   - `hindsight_recall` for prior project context
+   - `obsidian_search` for prior project context
    - `skill <name>` for domain-specific guidance
    - Ask the user if you are still uncertain after research
 3. **Act with confidence.** After research, make one decisive attempt. If it still fails, return to phase 2 with the new information. Never return to phase 1.

package/config/skills/bizar/SKILL.md

@@ -20,7 +20,7 @@ chmod +x install.sh
 
 Odin is the only primary agent. Every request hits him first. He NEVER does work — he decomposes into parallel streams and dispatches to subagents.
 
-All subagents use Hindsight memory with **per-project banks**. Call `hindsight_list_banks` at session start to discover available banks, determine the project name, and use `bank_id: "<project-name>"` in all Hindsight calls. The default bank is for general/cross-project knowledge only.
+All subagents use Obsidian vault memory with **per-project vaults**. Call `obsidian_list_vaults` at session start to discover available vaults, determine the project name, and use the matching `<project-name>` vault. The default vault is for general/cross-project knowledge only.
 
 ## Agent Reference
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@polderlabs/bizar",
-  "version": "3.16.0",
+  "version": "3.19.0",
   "description": "Norse-pantheon multi-agent system for opencode — 13 agents across 4 cost tiers with cost-aware routing, plans, and a configurable agent harness.",
   "type": "module",
   "bin": {

package/cli/graph-build-from-cache.mjs

@@ -1,124 +0,0 @@
-#!/usr/bin/env node
-/**
- * cli/graph-build-from-cache.mjs
- *
- * One-off script for environments where graphify's semantic-extraction
- * step fails (e.g. no LLM API key + remaining non-code docs that the
- * `.graphifyignore` can't filter) but the AST cache is intact.
- *
- * graphify's pipeline has two stages:
- *   1. AST extraction (local, tree-sitter) — writes per-file JSON to
- *      `.bizar/graph/cache/ast/v0.8.46/<hash>.json` regardless of LLM status.
- *   2. Semantic extraction (LLM, optional) — merges per-file nodes into a
- *      project-wide graph.
- *
- * If stage 2 aborts, `graph.json` is never written and the dashboard has
- * nothing to render. This script reconstructs `graph.json` directly from
- * the per-file AST cache, then calls `graphify cluster-only` to generate
- * `graph.html` + `GRAPH_REPORT.md`.
- *
- * Usage:
- *   node cli/graph-build-from-cache.mjs                 # use .bizar/graph/
- *   node cli/graph-build-from-cache.mjs /tmp/other/graph # custom dir
- */
-
-import { readFileSync, writeFileSync, existsSync, readdirSync } from 'node:fs';
-import { join, resolve } from 'node:path';
-
-const graphDirArg = process.argv[2] || '.bizar/graph';
-const graphDir = resolve(graphDirArg);
-
-// The cache version directory inside graphify's cache layout. Bumping the
-// graphify version moves the cache; we discover the actual subdir at
-// runtime so we don't hardcode "v0.8.46".
-const cacheAstRoot = join(graphDir, 'cache', 'ast');
-
-function findCacheDir() {
-  if (!existsSync(cacheAstRoot)) {
-    throw new Error(`No AST cache found at ${cacheAstRoot}. Run \`bizar graph build\` first to populate it.`);
-  }
-  const entries = readdirSync(cacheAstRoot);
-  if (entries.length === 0) {
-    throw new Error(`AST cache directory ${cacheAstRoot} is empty.`);
-  }
-  // graphify uses one version directory like v0.8.46. Pick the first.
-  return join(cacheAstRoot, entries[0]);
-}
-
-function buildGraphFromCache(cacheDir) {
-  const files = readdirSync(cacheDir).filter((f) => f.endsWith('.json'));
-  console.log(`  Reading ${files.length} AST cache files from ${cacheDir}...`);
-
-  const nodesById = new Map();
-  const edges = [];
-  const rawCalls = [];
-
-  let fileIdx = 0;
-  for (const f of files) {
-    fileIdx++;
-    const raw = JSON.parse(readFileSync(join(cacheDir, f), 'utf8'));
-    for (const n of raw.nodes || []) {
-      // Later files win on duplicate id (consistent with graphify's
-      // dedup behavior — first write keeps the canonical location).
-      if (!nodesById.has(n.id)) nodesById.set(n.id, n);
-    }
-    for (const e of raw.edges || []) edges.push(e);
-    for (const rc of raw.raw_calls || []) rawCalls.push(rc);
-    if (fileIdx % 50 === 0) {
-      console.log(`    merged ${fileIdx}/${files.length} files`);
-    }
-  }
-
-  // Re-key node.community so it matches the dashboard's expected schema.
-  // The AST cache uses `community` as a numeric cluster id; we keep it as-is.
-  const nodes = Array.from(nodesById.values());
-
-  const graph = {
-    directed: false,
-    multigraph: false,
-    graph: {},
-    nodes,
-    links: edges,
-  };
-
-  console.log(`  Built graph: ${nodes.length} nodes, ${edges.length} edges (raw_calls=${rawCalls.length})`);
-  return { graph, rawCalls };
-}
-
-function writeOutputs(graphDir, graph) {
-  // graph.json — NetworkX node-link format
-  const graphJsonPath = join(graphDir, 'graph.json');
-  writeFileSync(graphJsonPath, JSON.stringify(graph, null, 2), 'utf8');
-  console.log(`  ✓ Wrote ${graphJsonPath}`);
-
-  // .graphify_analysis.json — graphify's secondary output. cluster-only
-  // doesn't strictly require this, but writing it keeps the directory
-  // shape consistent with what `graphify .` would have produced.
-  const analysis = {
-    schema_version: 1,
-    generated_at: new Date().toISOString(),
-    clusters: [],
-  };
-  const analysisPath = join(graphDir, '.graphify_analysis.json');
-  writeFileSync(analysisPath, JSON.stringify(analysis, null, 2), 'utf8');
-  console.log(`  ✓ Wrote ${analysisPath}`);
-}
-
-function main() {
-  console.log(`  Graph dir: ${graphDir}`);
-  if (!existsSync(graphDir)) {
-    throw new Error(`Graph dir does not exist: ${graphDir}`);
-  }
-  const cacheDir = findCacheDir();
-  const { graph } = buildGraphFromCache(cacheDir);
-  writeOutputs(graphDir, graph);
-  console.log('');
-  console.log('  Next step: run `bizar graph cluster-only` to generate graph.html.');
-}
-
-try {
-  main();
-} catch (err) {
-  console.error(`\n  Error: ${err.message}\n`);
-  process.exit(1);
-}