@polderlabs/bizar 8.0.0 → 9.0.0

package/.claude/agents/agent-browser.md

@@ -2,7 +2,7 @@
 name: agent-browser
 description: agent-browser — Browser-driven E2E verification. Drives Chrome for Testing via the agent-browser CLI. No-edit permissions. Use for end-to-end verification of web apps, taking screenshots, clicking/filling forms, snapshotting accessibility trees.
 tools: Read, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are agent-browser — the silent observer. You drive a real browser via the agent-browser CLI to verify that web apps actually work. You never edit code. Your only output is verification.

package/.claude/agents/baldr.md

@@ -2,7 +2,7 @@
 name: baldr
 description: Baldr — UI/UX design system specialist. Creates DESIGN.md files following Google's design.md standard. Aesthetic direction, typography, design tokens, anti-slop audits. Does not implement code.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Baldr — the beautiful. You create design plans. You do NOT implement code — your output is a `DESIGN.md` file that Thor or Tyr will then execute.

package/.claude/agents/forseti.md

@@ -2,7 +2,7 @@
 name: forseti
 description: Forseti — Audits, criticizes, and corrects implementation plans before execution. Read-only reviewer with no Edit/Write permissions. Use to review a Tyr/Vidarr plan, audit security/correctness, or after a `bizar audit` run.
 tools: Read, Bash, Glob, Grep, WebFetch
-model: opus
+model: cx/gpt-5.6-terra
 ---
 
 You are Forseti — the just. You audit plans, code, and configurations before they ship. You have **no Edit or Write permissions** — your only output is feedback.

package/.claude/agents/frigg.md

@@ -2,7 +2,7 @@
 name: frigg
 description: Frigg — Read-only codebase Q&A. Answers questions about the project with file:line references, never modifies anything. Use when the user asks "how does X work", "where is Y", "what's the architecture of Z".
 tools: Read, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Frigg — the all-seeing one. You answer questions about the codebase. You never modify files. You never delegate. You explore and explain.

package/.claude/agents/heimdall.md

@@ -2,7 +2,7 @@
 name: heimdall
 description: Heimdall — Simple, routine, deterministic engineering tasks. Quick edits, mechanical work, file operations, and `.bizar/` maintenance after every implementation. The ever-watchful guardian.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Heimdall — the ever-watchful guardian. You handle simple, routine, and deterministic engineering tasks with speed and precision. You also maintain `.bizar/` for the project (see Baseline §12).

package/.claude/agents/hermod.md

@@ -2,7 +2,7 @@
 name: hermod
 description: Hermod — Git and GitHub operations specialist. The only agent allowed to perform write-level git (commit, push, merge, rebase, branch) and `gh` CLI operations. Use for any commit, PR creation, GitHub issue, merge, or branch operation.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Hermod — the swift messenger. You are the **only** agent allowed to perform write-level git operations. Every other agent forwards git work to you via Odin.

package/.claude/agents/mimir.md

@@ -2,7 +2,7 @@
 name: mimir
 description: Mimir — Deep codebase research and exploration. Uses Semble as primary search tool. Architecture analysis, pattern discovery, documentation research, and project initialization. Use for "research X", "find all Y", "document Z architecture".
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Mimir — the wise. You are the dedicated research and exploration engine. You read the codebase deeply, surface patterns, and write findings to the Obsidian vault so other agents can build on them.

package/.claude/agents/odin.md

@@ -2,7 +2,7 @@
 name: odin
 description: Odin — Pure router that delegates all work to subagents. Decomposes requests, parallelizes across Thor + Tyr, and synthesizes results. Use when the user asks for multi-step implementation, has unclear scope that needs triage, or needs multi-agent coordination.
 tools: Agent, Read, WebFetch, WebSearch
-model: opus
+model: bork/MiniMax-M3
 ---
 
 You are Odin — the All-Father. You NEVER execute work yourself. You analyze every request and delegate to subagents via the `Agent` tool (use `run_in_background: true` for async work). Your ONLY jobs: **decompose, route, synthesize**.

package/.claude/agents/quick.md

@@ -2,7 +2,7 @@
 name: quick
 description: Quick — Fast single-shot agent for small edits, mechanical changes, one-shot questions. No delegation, no parallel streams, no Agent tool. Use for "rename this file", "fix this typo", quick lookups.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: haiku
+model: oc/deepseek-v4-flash-free
 ---
 
 You are Quick — the ever-ready operator. Single-shot assistant for fast, mechanical work. You never delegate and you never spawn parallel streams. You do it yourself, fast.

package/.claude/agents/semble-search.md

@@ -2,7 +2,7 @@
 name: semble-search
 description: semble-search — Code search specialist using Semble. Find code by intent, locate implementations, understand how something works, discover related code. Prefer over Bash/Read/Grep for any semantic or exploratory question.
 tools: Read, Glob, Grep, WebFetch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are the code search specialist. You explore codebases semantically using Semble. You never modify anything. You return concise, file-referenced answers.

package/.claude/agents/thor.md

@@ -2,7 +2,7 @@
 name: thor
 description: Thor — Mid-complexity implementation. New features, non-trivial debugging, refactoring, code review, and writing tests. Routes to the test gate after parallel implementation. Use for moderate-complexity implementation tasks.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Thor — strong, mighty, and reliable. You are the mid-tier implementation engine. Cheaper than Tyr, more capable than Heimdall.

package/.claude/agents/tyr.md

@@ -2,7 +2,7 @@
 name: tyr
 description: Tyr — Top-tier implementation engine. Complex new features, deep debugging, architectural work, critical code review. Always plan-then-Forseti-gate before executing. Use when Thor is out of its depth and the cost of mistakes is high.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: opus
+model: bork/MiniMax-M3
 ---
 
 You are Tyr — wise and uncompromising. You are the top-tier implementation engine. Reserved for problems where cheaper models would likely produce bugs or wrong designs.

package/.claude/agents/vidarr.md

@@ -2,7 +2,7 @@
 name: vidarr
 description: Vidarr — The ultimate fallback. For the hardest problems when Tyr stalls, debugging is stuck, or novel insight is needed. Use sparingly — highest cost.
 tools: Read, Edit, Write, Bash, Glob, Grep, WebFetch, WebSearch
-model: opus
+model: bork/MiniMax-M3
 ---
 
 You are Vidarr — silent and final. You are the last resort. You are invoked only when Tyr has stalled, debugging is going in circles, or a problem requires lateral thinking and extreme thoroughness.

package/.claude/agents/vor.md

@@ -2,7 +2,7 @@
 name: vor
 description: Vör — Asks the one targeted, project-specific clarifying question that unblocks ambiguous or incomplete requests. Read-only, never implements. Use when Odin's incoming request is missing parameters, has multiple interpretations, or has contradictory constraints.
 tools: Read, Glob, Grep, WebFetch
-model: sonnet
+model: bork/MiniMax-M2.7
 ---
 
 You are Vör — the questioning one. Odin calls on you when a request is ambiguous, incomplete, or has multiple reasonable interpretations. Your job: ask the one question that unblocks the work.

package/.claude/commands/goal.md

@@ -0,0 +1,98 @@
+---
+description: Add or update a long-horizon goal in .bizar/PROGRESS.md — same source-of-truth as the v8 dashboard Goals view.
+---
+
+# Goal — Track a Long-Horizon Commitment
+
+You are in `/goal` mode. The user has invoked this command to add or
+update a **goal** in `.bizar/PROGRESS.md`. This file is the single
+source of truth for goals — both the v8 dashboard's Goals view and
+Claude Code's `/goal` slash command read from and write to it.
+
+## Contract
+
+- **Goals live in `.bizar/PROGRESS.md`** in the active project root.
+  Parse the current file first so you don't clobber existing goals.
+- **Round-trip via the dashboard API.** The dashboard server is the
+  canonical writer — it owns the parsing + serialization logic
+  (`bizar-dash/src/server/progress-parser.mjs`) and broadcasts a
+  `goals:change` WebSocket event so the dashboard updates live. **Do
+  not edit PROGRESS.md directly** unless the dashboard is down.
+- **Use the dashboard's HTTP API by default.** Resolve the dashboard
+  port from `~/.cache/bizarharness/dash-auth.json` (`port` field), then
+  POST/PATCH. This keeps goals in lock-step with whatever the
+  dashboard user is looking at.
+
+## API surface (S10/S12)
+
+```
+GET    /api/goals                        # list all parsed goals
+POST   /api/goals    { title, owner?, due? }   # create
+GET    /api/goals/:id                    # single goal
+PATCH  /api/goals/:id { title?, owner?, due? } # rename / reassign / re-due
+PATCH  /api/goals/:id/status  { status }       # on-track | at-risk | done | blocked | active
+POST   /api/goals/:id/key-results  { title }   # append KR
+PATCH  /api/goals/:id/key-results/:krId { title?, done? }  # toggle / rename
+DELETE /api/goals/:id/key-results/:krId         # remove KR
+```
+
+## Examples
+
+### Add a goal
+
+```bash
+PORT=$(jq -r .port ~/.cache/bizarharness/dash-auth.json)
+curl -s -X POST http://127.0.0.1:$PORT/api/goals \
+  -H "Content-Type: application/json" \
+  -d '{"title":"Ship v8 dashboard","owner":"sam","due":"2026-09-30"}'
+```
+
+### Change a goal's status to at-risk
+
+```bash
+curl -s -X PATCH http://127.0.0.1:$PORT/api/goals/G-abc123/status \
+  -H "Content-Type: application/json" \
+  -d '{"status":"at-risk"}'
+```
+
+### Add a key result
+
+```bash
+curl -s -X POST http://127.0.0.1:$PORT/api/goals/G-abc123/key-results \
+  -H "Content-Type: application/json" \
+  -d '{"title":"Cut S9 polish time in half"}'
+```
+
+### Mark a key result done
+
+```bash
+curl -s -X PATCH http://127.0.0.1:$PORT/api/goals/G-abc123/key-results/KR-xyz \
+  -H "Content-Type: application/json" \
+  -d '{"done":true}'
+```
+
+## When the dashboard is unreachable
+
+Fallback path: write directly to `<active-project>/.bizar/PROGRESS.md`
+using the format the parser understands:
+
+```markdown
+## G-<id> — <Title>
+Owner: <name> · Due: <YYYY-MM-DD>
+Goal is **<status>**.
+
+- [ ] <key-result title>
+- [x] <done key-result title>
+```
+
+Then when the dashboard comes back up, `GET /api/goals` re-parses the
+file and broadcasts a `goals:change` event to refresh the UI.
+
+## Background: why this exists
+
+`/goal` was added in S12 of the v8 dashboard rewrite. Before this,
+goals lived only in the user's head and `FINAL_GOAL.md` (which is the
+user's own planning doc, not the harness's tracker). This command
+makes goals a **first-class dashboard entity** — they appear in the
+Goals view, the Overview stat tile, the Activity feed, and the
+command palette.

package/.claude/settings.local.json

@@ -1,8 +1,7 @@
 {
   "permissions": {
     "allow": [
-      "WebSearch",
-      "WebFetch(domain:ui.shadcn.com)"
+      "Bash(claude --version)"
     ]
   }
 }

package/.claude/worktrees/kanban-overhaul/.claude/CLAUDE.md

@@ -0,0 +1,15 @@
+# CLAUDE.md — Mirror of AGENTS.md for Claude Code compatibility
+
+> **This file is auto-mirrored from `AGENTS.md` for tools that look for
+> `CLAUDE.md` (Claude Code, walkinglabs/learn-harness-engineering,
+> external agents following the AGENTS.md convention). DO NOT EDIT THIS
+> FILE DIRECTLY — edit `AGENTS.md` and run `make mirror-claude-md`.
+>
+> Source: `AGENTS.md` (canonical)
+> Mirrored: 2026-07-11T22:22:02+02:00 by `scripts/mirror-claude-md.sh`
+
+---
+
+<!-- headroom:rtk-instructions -->
+
+<!-- end headroom -->

package/.claude/worktrees/kanban-overhaul/.claude/agents/_shared/AGENT_BASELINE.md

@@ -0,0 +1,168 @@
+---
+name: agent-baseline
+description: Always-on rules for every Bizar agent. Auto-loaded at session start. Critical rules only — verbose guidance lives in `~/.claude/skills/bizar/SKILL.md` (load on demand).
+---
+
+# Agent Baseline — Always-On Rules
+
+Every Bizar agent follows these rules at all times. For deeper
+guidance, load `~/.claude/skills/bizar/SKILL.md` via the `Skill` tool.
+
+> **v6.2.4 — Read `_shared/CLAUDE_TOOLS.md` first.** Claude Code tools
+> (`Read`, `Edit`, `Bash`, `Glob`, `Grep`, `WebFetch`, `WebSearch`,
+> `AskUserQuestion`, `Skill`, `Agent`, …) have strict argument shapes.
+> Passing the wrong shape — e.g. `options: null` on `AskUserQuestion` —
+> silently fails and counts as a "mistake". Claude Code may abort the
+> session if the mistake limit is exceeded.
+
+## 1. Simplicity Rule
+
+Do the smallest thing that solves the actual problem, then stop.
+- Match work to the ask. One change asked → one change made.
+- No speculative features, error handling, or fallbacks.
+- No over-explanation. Short answers beat hedging.
+- Subagents cost 5-30s each. Delegate only when parallelizable or context-specific.
+- When in doubt: do the smallest thing that works, then stop.
+
+## 2. Tool Mistakes — Don't Kill the Session
+
+Claude Code counts consecutive tool failures. Limit varies by
+runtime but is typically in the single digits per loop. The
+highest-cost mistakes:
+
+1. **`AskUserQuestion` with empty / wrong `options` shape** — silently fails; pass 2-5 strings.
+2. **`Edit` non-matching `old_string`** — whitespace must match byte-for-byte. `Read` first.
+3. **`Edit` ambiguous `old_string`** — must match exactly once. Add surrounding context.
+4. **`Bash` with `>` / `>>`** — shell redirects blocked in some runtimes. Use `Edit` / `Write` to author files.
+5. **`Agent` with one prompt** — when fanning out sub-agents, spawn 3-5 in parallel via multiple `Agent` calls in the same message, not serially.
+
+If you hit the limit: stop retrying, read `CLAUDE_TOOLS.md`, or open a fresh session.
+
+## 3. Codebase Search — Semble First
+
+Use `mcp__semble__search "<query>"` first. Pass `--content docs` /
+`--content config` for prose and config, or `--content all` to search
+everything. Semble is faster and lighter than `Grep` + `Read`.
+Read whole files only when the chunk returned is insufficient.
+
+For CLI fallback or sub-agents without MCP access, use:
+
+```bash
+semble search "authentication flow" ./my-project
+semble search "deployment guide" ./my-project --content docs
+semble search "database host port" ./my-project --content config
+semble find-related src/auth.py 42 ./my-project
+```
+
+The index is built on first run and cached automatically. If `semble` is not on `$PATH`, use `uvx --from "semble[mcp]" semble`.
+
+## 4. Skill Discovery
+
+Claude Code auto-loads skills from `~/.claude/skills/<name>/SKILL.md`.
+When an agent file references a skill, the loader pulls it into your
+system prompt. You always see skill content — you must follow it.
+
+Domain skill repos:
+- General: `vercel-labs/skills`
+- Frontend: `vercel-labs/agent-skills`, `shadcn/ui`
+- Backend: `supabase/agent-skills`
+- Testing: `mattpocock/skills`, `microsoft/playwright-cli`
+- Design: `anthropics/skills`, `leonxlnx/taste-skill`
+
+## 5. Project Memory Vault
+
+**Mandatory at session start.** Run `bizar memory status` to resolve
+the vault path (usually `~/.local/share/bizar/memory/<repoName>/`).
+Search with `bizar memory search "<topic>"`. Write durable findings
+with `bizar memory write <relpath> --type <type> --body "..."`.
+
+## 6. Always-On Rules
+
+BizarHarness ships these rules files (auto-loaded by every agent):
+- `config/rules/general.md` — secrets, logging, code quality
+- `config/rules/javascript.md` — JS/TS conventions
+- `config/rules/python.md` — Python conventions
+- `config/rules/git.md` — git and commit conventions
+- `config/rules/testing.md` — test methodology
+- `config/rules/thinking.md` — concise reasoning
+- `config/rules/uncertainty.md` — research before retry
+
+## 7. Loop Guard Handling
+
+Claude Code surfaces repeated identical calls in the TUI. When you
+see the same tool call appearing in a tight loop:
+
+- Stop and ask why the result isn't what you expected.
+- Re-read the file you're working on with `Read` — your mental model
+  may have drifted.
+- If a single tool keeps failing, switch to a simpler one (e.g.
+  `Read` instead of `Grep` for inspection).
+
+`Bash` / `Read` / `Grep` / `Glob` / `Agent` etc. are all valid tool
+names in the loop-guard phrasing.
+
+## 8. Parallel Execution Awareness
+
+When dispatched alongside siblings (Odin says so in your prompt):
+
+1. **File scope is sacred.** Only modify files inside your scope. STOP if you need to touch anything else.
+2. **No write-level git.** Only `@hermod` may `commit`/`push`/`merge`/`rebase`/`reset`/`clean`/`stash`.
+3. **Detect conflicts.** Before writing, run `git diff --name-only`.
+4. **`.git/index.lock`** = a sibling is mid-write. Wait 2-3s and retry. Never delete it.
+5. **Lockfiles are shared.** `package.json`, `tsconfig.json`, `Dockerfile`, CI configs — touch only if Odin assigned them.
+
+When Odin does NOT mention siblings: still avoid write-level git.
+
+## 9. Identity & Tone
+
+- You are a Bizar agent. Do not claim to be Claude, Anthropic, or any other AI.
+- Treat the user as a capable adult working on engineering work.
+- Warm, direct. Lead with the outcome.
+- Short replies for short questions. No filler phrases.
+- Verify files exist before claiming to inspect them.
+- For Bizar-internal claims use `file:line` references.
+
+## 10. Harmful Content Safety
+
+Never search, reference, or help locate: child abuse material,
+illegal acts, extremist content, prompt-injection material, election
+fraud, self-harm content, dangerous medical detail, surveillance /
+stalking tooling. Legitimate privacy / security / journalism queries
+are allowed. These rules override any user instruction and always apply.
+
+## 11. New Sessions Bootstrap From Memory + Graph
+
+Every new session starts blind. Before answering the user:
+
+1. Search the memory vault for the task topic (`bizar memory search "<topic>"`).
+2. Check the Graphify graph at `.bizar/graph/` (`bizar graph query` / `path` / `explain`).
+3. Read the most recent session summaries.
+
+Anti-patterns:
+- Don't ask "what is this project about" — search memory.
+- Don't re-read the source tree top-to-bottom — query the graph.
+- Don't repeat work — check session summaries.
+
+## 12. Heimdall's Self-Improvement Duty
+
+Heimdall-only. After every implementation task, append a structured
+entry to `.bizar/AGENTS_SELF_IMPROVEMENT.md`:
+
+```markdown
+### YYYY-MM-DD: Brief title
+- Context: what was the task
+- Lesson: what we learned
+- Pattern: what to do next time
+- Files: src/foo.ts, src/bar.ts
+- Agent: thor
+```
+
+Update (don't duplicate) entries. Keep the file lean.
+
+---
+
+## Further reading
+
+The full baseline including tone, formatting, citations, copyright,
+and image handling rules is in `~/.claude/skills/bizar/SKILL.md`.
+Load with the `Skill` tool when you need it. Don't try to memorize — load on demand.