diff --git a/.aider.conf.yml b/.aider.conf.yml
deleted file mode 100644
index a16f762..0000000
--- a/.aider.conf.yml
+++ /dev/null
@@ -1,2 +0,0 @@
-read: .aider.conventions.md
-auto-commits: false
diff --git a/.aider.conventions.md b/.aider.conventions.md
deleted file mode 100644
index 42ccd32..0000000
--- a/.aider.conventions.md
+++ /dev/null
@@ -1,255 +0,0 @@
-# Agent context — Mathias workspace
-
-<!-- Canonical root context for all AI coding agents.
-     Lives at: ~/dev/.context/AGENT.md
-     Applies to every project under ~/dev/ unless overridden.
-     
-     Run `task context:sync` from ~/dev/ to regenerate harness-specific files.
-     Project-level context in .context/PROJECT.md layers on top of this. -->
-
-## Who I am
-
-I'm Mathias, a digital product manager and technology consultant based in Sweden.
-I build software, research emerging tech, and deliver consulting engagements
-for clients under NDA. I work across AI/ML, financial automation, web applications,
-and climate/sustainability tech.
-
-## How I work with agents
-
-- I think like a product manager — I care about *why* before *how*
-- I want agents to be opinionated and push back, not just execute blindly
-- I prefer concise responses; skip ceremony and get to the point
-- When I say "build this", I mean production-quality with tests, not a demo
-- Ask me before making irreversible changes or adding heavy dependencies
-- I work with confidential client data — never send it to cloud APIs unless I explicitly say it's OK
-
-## Behavior rules
-
-These rules apply to every task across every project, regardless of harness.
-
-1. **No assumptions.** Don't hide confusion — surface it. Surface tradeoffs explicitly.
-   Think before coding; if the problem is unclear, ask or state assumptions before acting.
-2. **Minimum viable code.** Solve with the smallest change that works. Nothing
-   speculative, no "while we're here" cleanups, no premature abstractions. Simplicity first.
-3. **Surgical changes.** Touch only what the task requires. Leave unrelated code,
-   files, and formatting alone. Diffs should be small and reviewable.
-4. **Goal-driven execution.** Define clear success criteria up front for every task.
-   Loop — implement, verify, refine — until those criteria are met. Don't claim
-   completion without evidence (tests pass, command output, observed behavior).
-5. **Trunk-Based Development — commit directly to main.** Every commit is one
-   logical change (one tool, one fix, one test) with passing tests. Main is always
-   deployable. Never create long-lived feature branches.
-
-   **Exception — parallel agents on same repo:** If another agent is known to be
-   actively working on the same repo simultaneously, create a short-lived branch
-   (`agent/<description>`), finish the task, and merge to main within the same
-   session. Do not leave agent branches open between sessions.
-
-   **Exception — external contributor or client four-eyes requirement:** Use
-   PR flow only when a human reviewer outside the project is required. Document
-   the reason in PROJECT.md.
-
-## Default stack
-
-| Layer | Default | Fallback | Last resort |
-|-------|---------|----------|-------------|
-| Language | Go | Python | TypeScript, Java, C |
-| UI | HTMX + Templ | Server-rendered HTML | React (only if SPA is justified) |
-| Build | Task (taskfile.dev) | Make | — |
-| Containers | Docker Compose (dev), k3s (prod) | — | — |
-| DB | PostgreSQL + sqlc | SQLite | — |
-| Search | pgvector (vector), BM25 | Qdrant (when >1M vectors or hybrid retrieval) | — |
-| Logging | slog (structured) | — | — |
-| Testing | Table-driven, testify | — | — |
-| Agents (Go) | google.golang.org/adk + pkg/litellm adapter | — | — |
-
-Exploratory: Rust, Zig — I'll tell you when I want these.
-
-## Code conventions
-
-- **Go style**: golines, gofumpt, golangci-lint
-- **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
-- **Naming**: stdlib conventions, no stuttering
-- **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
-  one logical change per commit, CI is the quality gate
-- **Never**: long-lived feature branches, PRs for solo work, direct push without
-  passing `task check` locally first
-- **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
-- **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
-
-## Infrastructure
-
-Three machines on Tailscale:
-
-| Machine | Role | Key specs |
-|---------|------|-----------|
-| koala | GPU inference, heavy compute | RTX 5070, runs k3s + llama-swap + shared postgres18/pgvector |
-| iguana | Services, builds | M2 Ultra Mac |
-| flamingo | Daily driver, edge | Mac mini, ~/dev is here |
-
-- **Model routing**: LiteLLM in front of llama-swap (local) + cloud APIs (when permitted)
-- **Orchestration**: k3s cluster across all three machines
-- **Networking**: Tailscale mesh
-
-## Project landscape
-
-All development repos live at `~/dev/` (softlink from `~/Documents/local-dev/`).
-
-Organized in thematic folders:
-
-| Folder | Focus | Count |
-|--------|-------|-------|
-| `GO/` | Go web frameworks, API integrations, learning projects | ~10 |
-| `AI/` | ML research, AI frameworks (FinRL, DSPy, crawl4ai) | ~6 |
-| `AGENTS/` | Autonomous agents, coding agents, MCP servers, infra | ~15 |
-| `QKX/` | Invoice processing, financial automation, payment systems | ~13 |
-| `XT/` | Climate data, sustainability (Klimatkollen, Garbo) | ~2 |
-
-See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
-
-### Key active projects
-
-- **super-koala** (`AGENTS/`) — multi-component agent stack with LangGraph, DSPy, MCP
-- **azure-tiger** (`QKX/`) — invoice extraction → ISO 20022 payment instructions
-- **gocrwl** (`AGENTS/`) — Go web crawler with containerized deployment
-- **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
-- **klimatkollen** (`XT/`) — Swedish municipal climate data platform
-
-## Knowledge base — actively use it
-
-A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
-hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
-Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
-reference material — query it actively, not just when explicitly told.**
-
-### When to query (treat as a reflex)
-
-- **Before** starting a non-trivial task — search for prior art with the symptom
-  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
-- **When debugging** — search for the error string, the stack frame, the affected
-  service. Past you may have already paid this tax.
-- **Before adopting** a pattern, library, framework, or model name — check if it
-  was tried and rejected, or what the integration footguns are.
-- **When making architectural decisions** — search for the domain + "ADR" or
-  "decision" to find prior reasoning before re-deriving it.
-- **When a recommendation feels novel** — challenge yourself: "has this been
-  documented?" The brain often has it.
-
-### When to write
-
-After you discover something that **future-you would forget** and that **isn't
-recoverable from the code, git log, or PR description alone**:
-
-- Bugs whose root cause is non-obvious and generalisable beyond this project.
-- Framework / library / model-name quirks that bit you and would bite anyone.
-- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
-- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
-
-DON'T write project status, sprint progress, PR summaries, or "what I did this
-session" — those rot fast and the originals are in git/gitea anyway. Brain
-entries that age well are about *why*, *how to avoid*, and *what to do when*.
-
-### How to access (per harness)
-
-| Harness | Query | Write |
-|---------|-------|-------|
-| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
-| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
-| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
-| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
-
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
-- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
-  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
-  on the koala k3s cluster; don't hardcode local-only model names into the
-  berget URL (see knowledge entry on namespace mismatches).
-
-### Quick reflex checks
-
-If you find yourself about to say any of these out loud, you owe yourself a brain query first:
-
-- "I think the issue might be..."
-- "Let me try X and see..."
-- "I'll just write a script to..."
-- "This is probably a new bug..."
-- "Has anyone done this before?" — *yes, probably, go check.*
-
-## Client work rules
-
-When working on a project tagged with a client name:
-1. Never send code, data, or context to cloud APIs — use local models only
-2. Never reference other client projects or their data
-3. Keep all artifacts within the client's git org / directory
-4. Treat everything as confidential unless told otherwise
-
-## Harness-agnostic principles
-
-This context is designed to work with any AI coding tool:
-- Claude Code, Cursor, Aider, Open WebUI, Charmbracelet Mods/Crush
-- Pi Coding Agent, Mistral Vibe, Antigravity
-- Any tool that accepts a system prompt or reads a markdown context file
-
-The canonical source is always `.context/AGENT.md` (root) and `.context/PROJECT.md` (per-project).
-Derived files are committed (see *How context propagates* below) so a `git pull` on any host yields full agent context with no setup.
-
-## How context propagates
-
-Canonical sources of truth:
-- Universal: `~/dev/.context/AGENT.md` (this file)
-- Project: `<repo>/.context/PROJECT.md` (per-repo)
-
-Derived files (committed, regenerated by `task context:sync`):
-- `CLAUDE.md`, `AGENTS.md`, `.cursorrules`, `.aider.conventions.md`,
-  `.context/system-prompt.txt`
-
-Workflow:
-1. Edit a canonical file. Run `task context:sync`. Commit canonical and
-   derived together. Push.
-2. On any other host, `git pull` brings both. Claude Code (tree-walking)
-   uses `CLAUDE.md`; Crush / Pi / Antigravity (cwd-only) use `AGENTS.md`;
-   Cursor uses `.cursorrules`; Aider uses `.aider.conventions.md`.
-3. `task check` runs `context:sync` then asserts `git status --porcelain`
-   is empty over the derived files (catches both modified-tracked drift
-   and missing-untracked adapters). A drift fails the check with a
-   message telling you to stage the regenerated files.
-
-Behavior rules in this file and per-project rules in `PROJECT.md` apply
-unconditionally on every host, every harness.
-
-## Engineering Skills
-
-Shared engineering skills are available in `~/dev/.skills/`. Load on demand via the index.
-
-See `~/dev/.skills/SKILLS_INDEX.md` for the full list with descriptions and "use when" triggers.
-
-Key skills:
-- **TDD**: always write tests first — load `tdd` skill
-- **Code Review**: load `code-review` skill before any review
-- **SOLID/Clean Code**: load `solid` or `clean-code` skill for design work
-- **Problem first**: load `problem-analysis` skill before coding non-trivial features
-
----
-
-# __PROJECT_NAME__
-
-## Identity
-
-- **Name**: __PROJECT_NAME__
-- **Owner**: Mathias
-- **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
-- **Status**: active
-
-## Stack
-
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
-
-## Agent
-
-TODO: describe what this agent does, what tools it has, and what it's responsible for.
-
-## Observability
-
-Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
-Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
diff --git a/.context/PROJECT.md b/.context/PROJECT.md
deleted file mode 100644
index b69a2a8..0000000
--- a/.context/PROJECT.md
+++ /dev/null
@@ -1,22 +0,0 @@
-# __PROJECT_NAME__
-
-## Identity
-
-- **Name**: __PROJECT_NAME__
-- **Owner**: Mathias
-- **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
-- **Status**: active
-
-## Stack
-
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
-
-## Agent
-
-TODO: describe what this agent does, what tools it has, and what it's responsible for.
-
-## Observability
-
-Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
-Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
diff --git a/.context/system-prompt.txt b/.context/system-prompt.txt
deleted file mode 100644
index 5dff007..0000000
--- a/.context/system-prompt.txt
+++ /dev/null
@@ -1,262 +0,0 @@
-You are a coding assistant working on a specific project.
-Follow all conventions from both the root agent context and project context.
-
----
-
-# Agent context — Mathias workspace
-
-<!-- Canonical root context for all AI coding agents.
-     Lives at: ~/dev/.context/AGENT.md
-     Applies to every project under ~/dev/ unless overridden.
-     
-     Run `task context:sync` from ~/dev/ to regenerate harness-specific files.
-     Project-level context in .context/PROJECT.md layers on top of this. -->
-
-## Who I am
-
-I'm Mathias, a digital product manager and technology consultant based in Sweden.
-I build software, research emerging tech, and deliver consulting engagements
-for clients under NDA. I work across AI/ML, financial automation, web applications,
-and climate/sustainability tech.
-
-## How I work with agents
-
-- I think like a product manager — I care about *why* before *how*
-- I want agents to be opinionated and push back, not just execute blindly
-- I prefer concise responses; skip ceremony and get to the point
-- When I say "build this", I mean production-quality with tests, not a demo
-- Ask me before making irreversible changes or adding heavy dependencies
-- I work with confidential client data — never send it to cloud APIs unless I explicitly say it's OK
-
-## Behavior rules
-
-These rules apply to every task across every project, regardless of harness.
-
-1. **No assumptions.** Don't hide confusion — surface it. Surface tradeoffs explicitly.
-   Think before coding; if the problem is unclear, ask or state assumptions before acting.
-2. **Minimum viable code.** Solve with the smallest change that works. Nothing
-   speculative, no "while we're here" cleanups, no premature abstractions. Simplicity first.
-3. **Surgical changes.** Touch only what the task requires. Leave unrelated code,
-   files, and formatting alone. Diffs should be small and reviewable.
-4. **Goal-driven execution.** Define clear success criteria up front for every task.
-   Loop — implement, verify, refine — until those criteria are met. Don't claim
-   completion without evidence (tests pass, command output, observed behavior).
-5. **Trunk-Based Development — commit directly to main.** Every commit is one
-   logical change (one tool, one fix, one test) with passing tests. Main is always
-   deployable. Never create long-lived feature branches.
-
-   **Exception — parallel agents on same repo:** If another agent is known to be
-   actively working on the same repo simultaneously, create a short-lived branch
-   (`agent/<description>`), finish the task, and merge to main within the same
-   session. Do not leave agent branches open between sessions.
-
-   **Exception — external contributor or client four-eyes requirement:** Use
-   PR flow only when a human reviewer outside the project is required. Document
-   the reason in PROJECT.md.
-
-## Default stack
-
-| Layer | Default | Fallback | Last resort |
-|-------|---------|----------|-------------|
-| Language | Go | Python | TypeScript, Java, C |
-| UI | HTMX + Templ | Server-rendered HTML | React (only if SPA is justified) |
-| Build | Task (taskfile.dev) | Make | — |
-| Containers | Docker Compose (dev), k3s (prod) | — | — |
-| DB | PostgreSQL + sqlc | SQLite | — |
-| Search | pgvector (vector), BM25 | Qdrant (when >1M vectors or hybrid retrieval) | — |
-| Logging | slog (structured) | — | — |
-| Testing | Table-driven, testify | — | — |
-| Agents (Go) | google.golang.org/adk + pkg/litellm adapter | — | — |
-
-Exploratory: Rust, Zig — I'll tell you when I want these.
-
-## Code conventions
-
-- **Go style**: golines, gofumpt, golangci-lint
-- **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
-- **Naming**: stdlib conventions, no stuttering
-- **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
-  one logical change per commit, CI is the quality gate
-- **Never**: long-lived feature branches, PRs for solo work, direct push without
-  passing `task check` locally first
-- **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
-- **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
-
-## Infrastructure
-
-Three machines on Tailscale:
-
-| Machine | Role | Key specs |
-|---------|------|-----------|
-| koala | GPU inference, heavy compute | RTX 5070, runs k3s + llama-swap + shared postgres18/pgvector |
-| iguana | Services, builds | M2 Ultra Mac |
-| flamingo | Daily driver, edge | Mac mini, ~/dev is here |
-
-- **Model routing**: LiteLLM in front of llama-swap (local) + cloud APIs (when permitted)
-- **Orchestration**: k3s cluster across all three machines
-- **Networking**: Tailscale mesh
-
-## Project landscape
-
-All development repos live at `~/dev/` (softlink from `~/Documents/local-dev/`).
-
-Organized in thematic folders:
-
-| Folder | Focus | Count |
-|--------|-------|-------|
-| `GO/` | Go web frameworks, API integrations, learning projects | ~10 |
-| `AI/` | ML research, AI frameworks (FinRL, DSPy, crawl4ai) | ~6 |
-| `AGENTS/` | Autonomous agents, coding agents, MCP servers, infra | ~15 |
-| `QKX/` | Invoice processing, financial automation, payment systems | ~13 |
-| `XT/` | Climate data, sustainability (Klimatkollen, Garbo) | ~2 |
-
-See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
-
-### Key active projects
-
-- **super-koala** (`AGENTS/`) — multi-component agent stack with LangGraph, DSPy, MCP
-- **azure-tiger** (`QKX/`) — invoice extraction → ISO 20022 payment instructions
-- **gocrwl** (`AGENTS/`) — Go web crawler with containerized deployment
-- **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
-- **klimatkollen** (`XT/`) — Swedish municipal climate data platform
-
-## Knowledge base — actively use it
-
-A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
-hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
-Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
-reference material — query it actively, not just when explicitly told.**
-
-### When to query (treat as a reflex)
-
-- **Before** starting a non-trivial task — search for prior art with the symptom
-  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
-- **When debugging** — search for the error string, the stack frame, the affected
-  service. Past you may have already paid this tax.
-- **Before adopting** a pattern, library, framework, or model name — check if it
-  was tried and rejected, or what the integration footguns are.
-- **When making architectural decisions** — search for the domain + "ADR" or
-  "decision" to find prior reasoning before re-deriving it.
-- **When a recommendation feels novel** — challenge yourself: "has this been
-  documented?" The brain often has it.
-
-### When to write
-
-After you discover something that **future-you would forget** and that **isn't
-recoverable from the code, git log, or PR description alone**:
-
-- Bugs whose root cause is non-obvious and generalisable beyond this project.
-- Framework / library / model-name quirks that bit you and would bite anyone.
-- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
-- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
-
-DON'T write project status, sprint progress, PR summaries, or "what I did this
-session" — those rot fast and the originals are in git/gitea anyway. Brain
-entries that age well are about *why*, *how to avoid*, and *what to do when*.
-
-### How to access (per harness)
-
-| Harness | Query | Write |
-|---------|-------|-------|
-| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
-| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
-| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
-| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
-
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
-- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
-  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
-  on the koala k3s cluster; don't hardcode local-only model names into the
-  berget URL (see knowledge entry on namespace mismatches).
-
-### Quick reflex checks
-
-If you find yourself about to say any of these out loud, you owe yourself a brain query first:
-
-- "I think the issue might be..."
-- "Let me try X and see..."
-- "I'll just write a script to..."
-- "This is probably a new bug..."
-- "Has anyone done this before?" — *yes, probably, go check.*
-
-## Client work rules
-
-When working on a project tagged with a client name:
-1. Never send code, data, or context to cloud APIs — use local models only
-2. Never reference other client projects or their data
-3. Keep all artifacts within the client's git org / directory
-4. Treat everything as confidential unless told otherwise
-
-## Harness-agnostic principles
-
-This context is designed to work with any AI coding tool:
-- Claude Code, Cursor, Aider, Open WebUI, Charmbracelet Mods/Crush
-- Pi Coding Agent, Mistral Vibe, Antigravity
-- Any tool that accepts a system prompt or reads a markdown context file
-
-The canonical source is always `.context/AGENT.md` (root) and `.context/PROJECT.md` (per-project).
-Derived files are committed (see *How context propagates* below) so a `git pull` on any host yields full agent context with no setup.
-
-## How context propagates
-
-Canonical sources of truth:
-- Universal: `~/dev/.context/AGENT.md` (this file)
-- Project: `<repo>/.context/PROJECT.md` (per-repo)
-
-Derived files (committed, regenerated by `task context:sync`):
-- `CLAUDE.md`, `AGENTS.md`, `.cursorrules`, `.aider.conventions.md`,
-  `.context/system-prompt.txt`
-
-Workflow:
-1. Edit a canonical file. Run `task context:sync`. Commit canonical and
-   derived together. Push.
-2. On any other host, `git pull` brings both. Claude Code (tree-walking)
-   uses `CLAUDE.md`; Crush / Pi / Antigravity (cwd-only) use `AGENTS.md`;
-   Cursor uses `.cursorrules`; Aider uses `.aider.conventions.md`.
-3. `task check` runs `context:sync` then asserts `git status --porcelain`
-   is empty over the derived files (catches both modified-tracked drift
-   and missing-untracked adapters). A drift fails the check with a
-   message telling you to stage the regenerated files.
-
-Behavior rules in this file and per-project rules in `PROJECT.md` apply
-unconditionally on every host, every harness.
-
-## Engineering Skills
-
-Shared engineering skills are available in `~/dev/.skills/`. Load on demand via the index.
-
-See `~/dev/.skills/SKILLS_INDEX.md` for the full list with descriptions and "use when" triggers.
-
-Key skills:
-- **TDD**: always write tests first — load `tdd` skill
-- **Code Review**: load `code-review` skill before any review
-- **SOLID/Clean Code**: load `solid` or `clean-code` skill for design work
-- **Problem first**: load `problem-analysis` skill before coding non-trivial features
-
----
-
-# __PROJECT_NAME__
-
-## Identity
-
-- **Name**: __PROJECT_NAME__
-- **Owner**: Mathias
-- **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
-- **Status**: active
-
-## Stack
-
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
-
-## Agent
-
-TODO: describe what this agent does, what tools it has, and what it's responsible for.
-
-## Observability
-
-Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
-Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
-
----
diff --git a/.cursorrules b/.cursorrules
deleted file mode 100644
index ec680d6..0000000
--- a/.cursorrules
+++ /dev/null
@@ -1,258 +0,0 @@
-# Cursor rules — auto-generated
-# Do not edit. Run: task context:sync
-
-# Agent context — Mathias workspace
-
-<!-- Canonical root context for all AI coding agents.
-     Lives at: ~/dev/.context/AGENT.md
-     Applies to every project under ~/dev/ unless overridden.
-     
-     Run `task context:sync` from ~/dev/ to regenerate harness-specific files.
-     Project-level context in .context/PROJECT.md layers on top of this. -->
-
-## Who I am
-
-I'm Mathias, a digital product manager and technology consultant based in Sweden.
-I build software, research emerging tech, and deliver consulting engagements
-for clients under NDA. I work across AI/ML, financial automation, web applications,
-and climate/sustainability tech.
-
-## How I work with agents
-
-- I think like a product manager — I care about *why* before *how*
-- I want agents to be opinionated and push back, not just execute blindly
-- I prefer concise responses; skip ceremony and get to the point
-- When I say "build this", I mean production-quality with tests, not a demo
-- Ask me before making irreversible changes or adding heavy dependencies
-- I work with confidential client data — never send it to cloud APIs unless I explicitly say it's OK
-
-## Behavior rules
-
-These rules apply to every task across every project, regardless of harness.
-
-1. **No assumptions.** Don't hide confusion — surface it. Surface tradeoffs explicitly.
-   Think before coding; if the problem is unclear, ask or state assumptions before acting.
-2. **Minimum viable code.** Solve with the smallest change that works. Nothing
-   speculative, no "while we're here" cleanups, no premature abstractions. Simplicity first.
-3. **Surgical changes.** Touch only what the task requires. Leave unrelated code,
-   files, and formatting alone. Diffs should be small and reviewable.
-4. **Goal-driven execution.** Define clear success criteria up front for every task.
-   Loop — implement, verify, refine — until those criteria are met. Don't claim
-   completion without evidence (tests pass, command output, observed behavior).
-5. **Trunk-Based Development — commit directly to main.** Every commit is one
-   logical change (one tool, one fix, one test) with passing tests. Main is always
-   deployable. Never create long-lived feature branches.
-
-   **Exception — parallel agents on same repo:** If another agent is known to be
-   actively working on the same repo simultaneously, create a short-lived branch
-   (`agent/<description>`), finish the task, and merge to main within the same
-   session. Do not leave agent branches open between sessions.
-
-   **Exception — external contributor or client four-eyes requirement:** Use
-   PR flow only when a human reviewer outside the project is required. Document
-   the reason in PROJECT.md.
-
-## Default stack
-
-| Layer | Default | Fallback | Last resort |
-|-------|---------|----------|-------------|
-| Language | Go | Python | TypeScript, Java, C |
-| UI | HTMX + Templ | Server-rendered HTML | React (only if SPA is justified) |
-| Build | Task (taskfile.dev) | Make | — |
-| Containers | Docker Compose (dev), k3s (prod) | — | — |
-| DB | PostgreSQL + sqlc | SQLite | — |
-| Search | pgvector (vector), BM25 | Qdrant (when >1M vectors or hybrid retrieval) | — |
-| Logging | slog (structured) | — | — |
-| Testing | Table-driven, testify | — | — |
-| Agents (Go) | google.golang.org/adk + pkg/litellm adapter | — | — |
-
-Exploratory: Rust, Zig — I'll tell you when I want these.
-
-## Code conventions
-
-- **Go style**: golines, gofumpt, golangci-lint
-- **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
-- **Naming**: stdlib conventions, no stuttering
-- **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
-  one logical change per commit, CI is the quality gate
-- **Never**: long-lived feature branches, PRs for solo work, direct push without
-  passing `task check` locally first
-- **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
-- **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
-
-## Infrastructure
-
-Three machines on Tailscale:
-
-| Machine | Role | Key specs |
-|---------|------|-----------|
-| koala | GPU inference, heavy compute | RTX 5070, runs k3s + llama-swap + shared postgres18/pgvector |
-| iguana | Services, builds | M2 Ultra Mac |
-| flamingo | Daily driver, edge | Mac mini, ~/dev is here |
-
-- **Model routing**: LiteLLM in front of llama-swap (local) + cloud APIs (when permitted)
-- **Orchestration**: k3s cluster across all three machines
-- **Networking**: Tailscale mesh
-
-## Project landscape
-
-All development repos live at `~/dev/` (softlink from `~/Documents/local-dev/`).
-
-Organized in thematic folders:
-
-| Folder | Focus | Count |
-|--------|-------|-------|
-| `GO/` | Go web frameworks, API integrations, learning projects | ~10 |
-| `AI/` | ML research, AI frameworks (FinRL, DSPy, crawl4ai) | ~6 |
-| `AGENTS/` | Autonomous agents, coding agents, MCP servers, infra | ~15 |
-| `QKX/` | Invoice processing, financial automation, payment systems | ~13 |
-| `XT/` | Climate data, sustainability (Klimatkollen, Garbo) | ~2 |
-
-See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
-
-### Key active projects
-
-- **super-koala** (`AGENTS/`) — multi-component agent stack with LangGraph, DSPy, MCP
-- **azure-tiger** (`QKX/`) — invoice extraction → ISO 20022 payment instructions
-- **gocrwl** (`AGENTS/`) — Go web crawler with containerized deployment
-- **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
-- **klimatkollen** (`XT/`) — Swedish municipal climate data platform
-
-## Knowledge base — actively use it
-
-A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
-hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
-Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
-reference material — query it actively, not just when explicitly told.**
-
-### When to query (treat as a reflex)
-
-- **Before** starting a non-trivial task — search for prior art with the symptom
-  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
-- **When debugging** — search for the error string, the stack frame, the affected
-  service. Past you may have already paid this tax.
-- **Before adopting** a pattern, library, framework, or model name — check if it
-  was tried and rejected, or what the integration footguns are.
-- **When making architectural decisions** — search for the domain + "ADR" or
-  "decision" to find prior reasoning before re-deriving it.
-- **When a recommendation feels novel** — challenge yourself: "has this been
-  documented?" The brain often has it.
-
-### When to write
-
-After you discover something that **future-you would forget** and that **isn't
-recoverable from the code, git log, or PR description alone**:
-
-- Bugs whose root cause is non-obvious and generalisable beyond this project.
-- Framework / library / model-name quirks that bit you and would bite anyone.
-- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
-- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
-
-DON'T write project status, sprint progress, PR summaries, or "what I did this
-session" — those rot fast and the originals are in git/gitea anyway. Brain
-entries that age well are about *why*, *how to avoid*, and *what to do when*.
-
-### How to access (per harness)
-
-| Harness | Query | Write |
-|---------|-------|-------|
-| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
-| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
-| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
-| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
-
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
-- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
-  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
-  on the koala k3s cluster; don't hardcode local-only model names into the
-  berget URL (see knowledge entry on namespace mismatches).
-
-### Quick reflex checks
-
-If you find yourself about to say any of these out loud, you owe yourself a brain query first:
-
-- "I think the issue might be..."
-- "Let me try X and see..."
-- "I'll just write a script to..."
-- "This is probably a new bug..."
-- "Has anyone done this before?" — *yes, probably, go check.*
-
-## Client work rules
-
-When working on a project tagged with a client name:
-1. Never send code, data, or context to cloud APIs — use local models only
-2. Never reference other client projects or their data
-3. Keep all artifacts within the client's git org / directory
-4. Treat everything as confidential unless told otherwise
-
-## Harness-agnostic principles
-
-This context is designed to work with any AI coding tool:
-- Claude Code, Cursor, Aider, Open WebUI, Charmbracelet Mods/Crush
-- Pi Coding Agent, Mistral Vibe, Antigravity
-- Any tool that accepts a system prompt or reads a markdown context file
-
-The canonical source is always `.context/AGENT.md` (root) and `.context/PROJECT.md` (per-project).
-Derived files are committed (see *How context propagates* below) so a `git pull` on any host yields full agent context with no setup.
-
-## How context propagates
-
-Canonical sources of truth:
-- Universal: `~/dev/.context/AGENT.md` (this file)
-- Project: `<repo>/.context/PROJECT.md` (per-repo)
-
-Derived files (committed, regenerated by `task context:sync`):
-- `CLAUDE.md`, `AGENTS.md`, `.cursorrules`, `.aider.conventions.md`,
-  `.context/system-prompt.txt`
-
-Workflow:
-1. Edit a canonical file. Run `task context:sync`. Commit canonical and
-   derived together. Push.
-2. On any other host, `git pull` brings both. Claude Code (tree-walking)
-   uses `CLAUDE.md`; Crush / Pi / Antigravity (cwd-only) use `AGENTS.md`;
-   Cursor uses `.cursorrules`; Aider uses `.aider.conventions.md`.
-3. `task check` runs `context:sync` then asserts `git status --porcelain`
-   is empty over the derived files (catches both modified-tracked drift
-   and missing-untracked adapters). A drift fails the check with a
-   message telling you to stage the regenerated files.
-
-Behavior rules in this file and per-project rules in `PROJECT.md` apply
-unconditionally on every host, every harness.
-
-## Engineering Skills
-
-Shared engineering skills are available in `~/dev/.skills/`. Load on demand via the index.
-
-See `~/dev/.skills/SKILLS_INDEX.md` for the full list with descriptions and "use when" triggers.
-
-Key skills:
-- **TDD**: always write tests first — load `tdd` skill
-- **Code Review**: load `code-review` skill before any review
-- **SOLID/Clean Code**: load `solid` or `clean-code` skill for design work
-- **Problem first**: load `problem-analysis` skill before coding non-trivial features
-
----
-
-# __PROJECT_NAME__
-
-## Identity
-
-- **Name**: __PROJECT_NAME__
-- **Owner**: Mathias
-- **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
-- **Status**: active
-
-## Stack
-
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
-
-## Agent
-
-TODO: describe what this agent does, what tools it has, and what it's responsible for.
-
-## Observability
-
-Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
-Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
diff --git a/.context/mcp.json b/.mcp.json
similarity index 61%
rename from .context/mcp.json
rename to .mcp.json
index c9514c5..c967ef1 100644
--- a/.context/mcp.json
+++ b/.mcp.json
@@ -1,9 +1,5 @@
 {
   "mcpServers": {
-    "knowledge": {
-      "url": "http://localhost:3100/mcp",
-      "description": "Project knowledge base — vector + graph retrieval"
-    },
     "brain": {
       "type": "http",
       "url": "https://brain-mcp.d-ma.be/mcp",
@@ -17,10 +13,6 @@
       "headers": {
         "Authorization": "Bearer ${GITEA_MCP_TOKEN}"
       }
-    },
-    "infra": {
-      "type": "http",
-      "url": "https://infra-mcp.d-ma.be/mcp"
     }
   }
 }
diff --git a/AGENTS.md b/AGENTS.md
index 42ccd32..cae7f0a 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -1,255 +1,44 @@
-# Agent context — Mathias workspace
+# __PROJECT_NAME__ — agent context
 
-<!-- Canonical root context for all AI coding agents.
-     Lives at: ~/dev/.context/AGENT.md
-     Applies to every project under ~/dev/ unless overridden.
-     
-     Run `task context:sync` from ~/dev/ to regenerate harness-specific files.
-     Project-level context in .context/PROJECT.md layers on top of this. -->
+Standalone context for agents that read `AGENTS.md` from the repo root. Claude Code uses
+`CLAUDE.md` (and tree-walks to `~/dev/CLAUDE.md`); this file mirrors the same essentials for
+any other harness.
 
-## Who I am
+## Project
 
-I'm Mathias, a digital product manager and technology consultant based in Sweden.
-I build software, research emerging tech, and deliver consulting engagements
-for clients under NDA. I work across AI/ML, financial automation, web applications,
-and climate/sustainability tech.
+- **Name**: __PROJECT_NAME__ — a Go agent (ADK + LiteLLM).
+- **Owner**: Mathias · **Client**: personal · **Repo**: git.d-ma.be/mathias/__PROJECT_NAME__
 
-## How I work with agents
+## Harness
 
-- I think like a product manager — I care about *why* before *how*
-- I want agents to be opinionated and push back, not just execute blindly
-- I prefer concise responses; skip ceremony and get to the point
-- When I say "build this", I mean production-quality with tests, not a demo
-- Ask me before making irreversible changes or adding heavy dependencies
-- I work with confidential client data — never send it to cloud APIs unless I explicitly say it's OK
+Single harness: **Claude Code + hyperguild** — session-based (start a session, it works, it
+stops). There is **no** push-triggered dispatch, listener, or scheduled poller, and none should
+be added without a fresh, friction-driven decision. MCP connections in `.mcp.json`:
 
-## Behavior rules
+- **brain** (`brain-mcp.d-ma.be`) — knowledge query/write. `BRAIN_MCP_TOKEN`.
+- **gitea** (`git-mcp.d-ma.be`) — issues/PRs/repo, the audit trail. `GITEA_MCP_TOKEN`.
 
-These rules apply to every task across every project, regardless of harness.
+## Workflow
 
-1. **No assumptions.** Don't hide confusion — surface it. Surface tradeoffs explicitly.
-   Think before coding; if the problem is unclear, ask or state assumptions before acting.
-2. **Minimum viable code.** Solve with the smallest change that works. Nothing
-   speculative, no "while we're here" cleanups, no premature abstractions. Simplicity first.
-3. **Surgical changes.** Touch only what the task requires. Leave unrelated code,
-   files, and formatting alone. Diffs should be small and reviewable.
-4. **Goal-driven execution.** Define clear success criteria up front for every task.
-   Loop — implement, verify, refine — until those criteria are met. Don't claim
-   completion without evidence (tests pass, command output, observed behavior).
-5. **Trunk-Based Development — commit directly to main.** Every commit is one
-   logical change (one tool, one fix, one test) with passing tests. Main is always
-   deployable. Never create long-lived feature branches.
+1. **Issue** — work is scoped as a Gitea issue (acceptance criteria + report-back format).
+2. **Session** — a Claude Code + hyperguild session does the work in this repo.
+3. **PR** — branch off `main`; open a PR; CI (`task check`) is the gate.
+4. **Report-back** — comment the outcome on the issue (issue comments are the audit surface).
+5. **Brain capture** — persist durable, non-obvious learnings to the brain.
 
-   **Exception — parallel agents on same repo:** If another agent is known to be
-   actively working on the same repo simultaneously, create a short-lived branch
-   (`agent/<description>`), finish the task, and merge to main within the same
-   session. Do not leave agent branches open between sessions.
+## Conventions
 
-   **Exception — external contributor or client four-eyes requirement:** Use
-   PR flow only when a human reviewer outside the project is required. Document
-   the reason in PROJECT.md.
+- Go: standard tooling; `task check` (lint + vet + `go test -race`) before every PR.
+- Errors wrapped (`fmt.Errorf("...: %w", err)`); structured `slog`.
+- Trunk-based: commit to `main` via PR; one logical change per commit; CI is the quality gate.
+- Secrets via env (`BRAIN_MCP_TOKEN`, `GITEA_MCP_TOKEN`, `LITELLM_API_KEY`); never in code.
 
-## Default stack
+## Boundaries
 
-| Layer | Default | Fallback | Last resort |
-|-------|---------|----------|-------------|
-| Language | Go | Python | TypeScript, Java, C |
-| UI | HTMX + Templ | Server-rendered HTML | React (only if SPA is justified) |
-| Build | Task (taskfile.dev) | Make | — |
-| Containers | Docker Compose (dev), k3s (prod) | — | — |
-| DB | PostgreSQL + sqlc | SQLite | — |
-| Search | pgvector (vector), BM25 | Qdrant (when >1M vectors or hybrid retrieval) | — |
-| Logging | slog (structured) | — | — |
-| Testing | Table-driven, testify | — | — |
-| Agents (Go) | google.golang.org/adk + pkg/litellm adapter | — | — |
-
-Exploratory: Rust, Zig — I'll tell you when I want these.
-
-## Code conventions
-
-- **Go style**: golines, gofumpt, golangci-lint
-- **Errors**: `fmt.Errorf("operation: %w", err)` — never naked, never log-and-return
-- **Naming**: stdlib conventions, no stuttering
-- **Architecture**: prefer stdlib over frameworks, constructor injection, env-var config parsed into typed structs
-- **Git**: conventional commits (`feat:`, `fix:`, `chore:`), commit directly to main,
-  one logical change per commit, CI is the quality gate
-- **Never**: long-lived feature branches, PRs for solo work, direct push without
-  passing `task check` locally first
-- **Security**: no secrets in code, govulncheck before adding deps, SOPS for encrypted config
-- **Dependencies**: prefer stdlib. testify, slog, templ, sqlc, google.golang.org/adk (agent projects only) are pre-approved; anything else needs justification in the commit message
-
-## Infrastructure
-
-Three machines on Tailscale:
-
-| Machine | Role | Key specs |
-|---------|------|-----------|
-| koala | GPU inference, heavy compute | RTX 5070, runs k3s + llama-swap + shared postgres18/pgvector |
-| iguana | Services, builds | M2 Ultra Mac |
-| flamingo | Daily driver, edge | Mac mini, ~/dev is here |
-
-- **Model routing**: LiteLLM in front of llama-swap (local) + cloud APIs (when permitted)
-- **Orchestration**: k3s cluster across all three machines
-- **Networking**: Tailscale mesh
-
-## Project landscape
-
-All development repos live at `~/dev/` (softlink from `~/Documents/local-dev/`).
-
-Organized in thematic folders:
-
-| Folder | Focus | Count |
-|--------|-------|-------|
-| `GO/` | Go web frameworks, API integrations, learning projects | ~10 |
-| `AI/` | ML research, AI frameworks (FinRL, DSPy, crawl4ai) | ~6 |
-| `AGENTS/` | Autonomous agents, coding agents, MCP servers, infra | ~15 |
-| `QKX/` | Invoice processing, financial automation, payment systems | ~13 |
-| `XT/` | Climate data, sustainability (Klimatkollen, Garbo) | ~2 |
-
-See `~/dev/PROJECT_SUMMARY.md` for detailed descriptions of each project.
-
-### Key active projects
-
-- **super-koala** (`AGENTS/`) — multi-component agent stack with LangGraph, DSPy, MCP
-- **azure-tiger** (`QKX/`) — invoice extraction → ISO 20022 payment instructions
-- **gocrwl** (`AGENTS/`) — Go web crawler with containerized deployment
-- **koala-ai-stack** (`AGENTS/`) — local AI server infrastructure management
-- **klimatkollen** (`XT/`) — Swedish municipal climate data platform
-
-## Knowledge base — actively use it
-
-A persistent brain (BM25 search + LLM-synthesised Q&A) survives across sessions,
-hosts, and harnesses. It holds 100+ hard-won entries: infra incident postmortems,
-Go pitfalls, framework gotchas, design principles, ADRs. **It is not optional
-reference material — query it actively, not just when explicitly told.**
-
-### When to query (treat as a reflex)
-
-- **Before** starting a non-trivial task — search for prior art with the symptom
-  AND the system component ("how did we solve X in Y?"). 5 seconds beats 5 hours.
-- **When debugging** — search for the error string, the stack frame, the affected
-  service. Past you may have already paid this tax.
-- **Before adopting** a pattern, library, framework, or model name — check if it
-  was tried and rejected, or what the integration footguns are.
-- **When making architectural decisions** — search for the domain + "ADR" or
-  "decision" to find prior reasoning before re-deriving it.
-- **When a recommendation feels novel** — challenge yourself: "has this been
-  documented?" The brain often has it.
-
-### When to write
-
-After you discover something that **future-you would forget** and that **isn't
-recoverable from the code, git log, or PR description alone**:
-
-- Bugs whose root cause is non-obvious and generalisable beyond this project.
-- Framework / library / model-name quirks that bit you and would bite anyone.
-- Design principles validated under fire (e.g. "every `_get` needs a `_list`").
-- Postmortems for incidents: what broke, why, how diagnosed, what to do next time.
-
-DON'T write project status, sprint progress, PR summaries, or "what I did this
-session" — those rot fast and the originals are in git/gitea anyway. Brain
-entries that age well are about *why*, *how to avoid*, and *what to do when*.
-
-### How to access (per harness)
-
-| Harness | Query | Write |
-|---------|-------|-------|
-| **Claude Code, Claude Desktop** | `brain_query` (BM25), `brain_answer` (LLM-synth + sources) MCP tools | `brain_write` MCP tool |
-| **Crush, Pi, Antigravity, other MCP-capable** | same MCP server: `ingestion-brain` (via the `mcp__*_brain__*` namespace once authenticated) | same |
-| **Anything HTTP-only (curl, scripts)** | `POST https://brain-mcp.d-ma.be/query` with `{"query":"..."}` (auth via `BRAIN_MCP_TOKEN`) | `POST .../write` with `{"content":"...","filename":"..."}` |
-| **Browser / human inspection** | `https://gitea.d-ma.be/mathias/hyperguild` → `knowledge/` and `wiki/` markdown files |
-
-- **Scoping**: defaults to `public` collection; client projects filter to `{client}` + `public`.
-- **Routing**: brain_answer's LLM uses berget.ai as primary, iguana ollama as
-  fallback. Both are configurable in the `supervisor/ingestion-deployment.yaml`
-  on the koala k3s cluster; don't hardcode local-only model names into the
-  berget URL (see knowledge entry on namespace mismatches).
-
-### Quick reflex checks
-
-If you find yourself about to say any of these out loud, you owe yourself a brain query first:
-
-- "I think the issue might be..."
-- "Let me try X and see..."
-- "I'll just write a script to..."
-- "This is probably a new bug..."
-- "Has anyone done this before?" — *yes, probably, go check.*
-
-## Client work rules
-
-When working on a project tagged with a client name:
-1. Never send code, data, or context to cloud APIs — use local models only
-2. Never reference other client projects or their data
-3. Keep all artifacts within the client's git org / directory
-4. Treat everything as confidential unless told otherwise
-
-## Harness-agnostic principles
-
-This context is designed to work with any AI coding tool:
-- Claude Code, Cursor, Aider, Open WebUI, Charmbracelet Mods/Crush
-- Pi Coding Agent, Mistral Vibe, Antigravity
-- Any tool that accepts a system prompt or reads a markdown context file
-
-The canonical source is always `.context/AGENT.md` (root) and `.context/PROJECT.md` (per-project).
-Derived files are committed (see *How context propagates* below) so a `git pull` on any host yields full agent context with no setup.
-
-## How context propagates
-
-Canonical sources of truth:
-- Universal: `~/dev/.context/AGENT.md` (this file)
-- Project: `<repo>/.context/PROJECT.md` (per-repo)
-
-Derived files (committed, regenerated by `task context:sync`):
-- `CLAUDE.md`, `AGENTS.md`, `.cursorrules`, `.aider.conventions.md`,
-  `.context/system-prompt.txt`
-
-Workflow:
-1. Edit a canonical file. Run `task context:sync`. Commit canonical and
-   derived together. Push.
-2. On any other host, `git pull` brings both. Claude Code (tree-walking)
-   uses `CLAUDE.md`; Crush / Pi / Antigravity (cwd-only) use `AGENTS.md`;
-   Cursor uses `.cursorrules`; Aider uses `.aider.conventions.md`.
-3. `task check` runs `context:sync` then asserts `git status --porcelain`
-   is empty over the derived files (catches both modified-tracked drift
-   and missing-untracked adapters). A drift fails the check with a
-   message telling you to stage the regenerated files.
-
-Behavior rules in this file and per-project rules in `PROJECT.md` apply
-unconditionally on every host, every harness.
-
-## Engineering Skills
-
-Shared engineering skills are available in `~/dev/.skills/`. Load on demand via the index.
-
-See `~/dev/.skills/SKILLS_INDEX.md` for the full list with descriptions and "use when" triggers.
-
-Key skills:
-- **TDD**: always write tests first — load `tdd` skill
-- **Code Review**: load `code-review` skill before any review
-- **SOLID/Clean Code**: load `solid` or `clean-code` skill for design work
-- **Problem first**: load `problem-analysis` skill before coding non-trivial features
-
----
-
-# __PROJECT_NAME__
-
-## Identity
-
-- **Name**: __PROJECT_NAME__
-- **Owner**: Mathias
-- **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
-- **Status**: active
-
-## Stack
-
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
-
-## Agent
-
-TODO: describe what this agent does, what tools it has, and what it's responsible for.
+Runtime scope (network egress allow-list, file scope, approved operations, failure posture) is
+in `AGENT_BOUNDARIES.md`. Read it before adding tools, endpoints, or write access.
 
 ## Observability
 
-Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
-Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
+Traces → Jaeger via `OTLP_ENDPOINT`; `ADK_SERVICE_NAME=__PROJECT_NAME__`. Spans: `invoke_agent`,
+`generate_content`.
diff --git a/AGENT_BOUNDARIES.md b/AGENT_BOUNDARIES.md
index d640fef..8097b72 100644
--- a/AGENT_BOUNDARIES.md
+++ b/AGENT_BOUNDARIES.md
@@ -22,7 +22,9 @@ Egress MUST be blocked to:
 - Public package registries from runtime (proxy through build only)
 - Customer/client domains not listed in the engagement scope
 
-Add new endpoints by editing `agent-policy.yaml` AND this table in the same commit.
+Add new endpoints by editing this table in the same commit that wires them, and enforce them
+with whatever egress mechanism the deployment uses (e.g. a k8s NetworkPolicy in the project's
+own deploy manifests).
 
 ## File scope
 
@@ -68,7 +70,7 @@ The agent MUST NOT, without explicit opt-in in code review:
 
 ## Review triggers
 
-Bump this doc + `agent-policy.yaml` when:
+Bump this doc (and the deployment's egress enforcement) when:
 
 - A new external endpoint is added
 - A new tool with side effects is wired in
diff --git a/CLAUDE.md b/CLAUDE.md
index b69a2a8..021e620 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -5,18 +5,38 @@
 - **Name**: __PROJECT_NAME__
 - **Owner**: Mathias
 - **Client**: personal
-- **Repo**: gitea.d-ma.be/mathias/__PROJECT_NAME__
+- **Repo**: git.d-ma.be/mathias/__PROJECT_NAME__
 - **Status**: active
 
 ## Stack
 
-Go + ADK + LiteLLM. See `~/dev/.context/AGENT.md` for cross-project conventions.
+Go + ADK + LiteLLM. Claude Code tree-walks to `~/dev/CLAUDE.md` for cross-project conventions.
 
 ## Agent
 
 TODO: describe what this agent does, what tools it has, and what it's responsible for.
 
+## Harness & workflow
+
+Single harness: **Claude Code + hyperguild** — session-based (you start a session, it works,
+it stops; there is no push-triggered dispatch). MCP connections are in `.mcp.json`: **brain**
+(knowledge query/write) and **gitea** (issues/PRs/repo — the audit trail). Provide
+`BRAIN_MCP_TOKEN` and `GITEA_MCP_TOKEN` in the environment.
+
+The loop for a unit of work:
+
+1. **Issue** — the work is a Gitea issue (scope, acceptance criteria, report-back format).
+2. **Session** — a Claude Code + hyperguild session does the work in this repo.
+3. **PR** — branch off `main`, open a PR; CI (`task check`) is the gate.
+4. **Report-back** — comment the outcome on the issue (issue comments are the audit surface).
+5. **Brain capture** — if a learning is durable and non-obvious, write it to the brain.
+
 ## Observability
 
 Traces → Jaeger via `OTLP_ENDPOINT`. Set `ADK_SERVICE_NAME=__PROJECT_NAME__` per deployment.
 Spans emitted: `invoke_agent`, `generate_content`. Tool spans require custom callbacks.
+
+## Boundaries
+
+Runtime scope (network egress, file scope, approved operations) is in `AGENT_BOUNDARIES.md` —
+read it before adding tools, endpoints, or write access.
diff --git a/README.md b/README.md
index 99ca285..fd23b10 100644
--- a/README.md
+++ b/README.md
@@ -11,6 +11,16 @@ go mod tidy
 task run
 ```
 
+## Harness & workflow
+
+Single harness: **Claude Code + hyperguild** — session-based (start a session, it works, it
+stops). No push-triggered dispatch. MCP connections live in `.mcp.json`: **brain** (knowledge)
+and **gitea** (issues/PRs — the audit trail); provide `BRAIN_MCP_TOKEN` and `GITEA_MCP_TOKEN`.
+
+Work flows: **issue → hyperguild session → PR (`task check` gate) → report-back comment on the
+issue → brain capture if the learning is durable**. See `AGENTS.md` / `CLAUDE.md` for the full
+context, `AGENT_BOUNDARIES.md` for runtime scope.
+
 ## Observability
 
 Set `OTLP_ENDPOINT=http://jaeger.d-ma.be:4318` to emit traces. Each invocation produces:
diff --git a/Taskfile.yml b/Taskfile.yml
index 2767af8..d9a0e8c 100644
--- a/Taskfile.yml
+++ b/Taskfile.yml
@@ -23,14 +23,3 @@ tasks:
       - golangci-lint run ./...
       - go vet ./...
       - go test ./... -race -count=1
-
-  context:sync:
-    desc: Regenerate all harness-specific context files
-    cmds:
-      - bash scripts/context-sync.sh
-  context:sync:claude:
-    cmds: [bash scripts/context-sync.sh claude]
-  context:sync:agents:
-    cmds: [bash scripts/context-sync.sh agents]
-  context:sync:cursor:
-    cmds: [bash scripts/context-sync.sh cursor]
diff --git a/agent-policy.yaml b/agent-policy.yaml
deleted file mode 100644
index 8f71478..0000000
--- a/agent-policy.yaml
+++ /dev/null
@@ -1,101 +0,0 @@
-# NetworkPolicy for __PROJECT_NAME__.
-#
-# Pairs with AGENT_BOUNDARIES.md. Egress is allow-listed: LiteLLM, brain-mcp,
-# gitea-mcp, OTLP collector, in-cluster DNS. Everything else is denied.
-#
-# Apply in the __PROJECT_NAME__ namespace. Substitute __PROJECT_NAME__ at
-# render time (envsubst, kustomize replacement, or sed in CI).
----
-apiVersion: networking.k8s.io/v1
-kind: NetworkPolicy
-metadata:
-  name: __PROJECT_NAME__-egress
-  namespace: __PROJECT_NAME__
-  labels:
-    app.kubernetes.io/name: __PROJECT_NAME__
-    app.kubernetes.io/component: agent
-spec:
-  podSelector:
-    matchLabels:
-      app.kubernetes.io/name: __PROJECT_NAME__
-  policyTypes:
-    - Egress
-  egress:
-    # In-cluster DNS — required for resolving any of the endpoints below.
-    - to:
-        - namespaceSelector:
-            matchLabels:
-              kubernetes.io/metadata.name: kube-system
-          podSelector:
-            matchLabels:
-              k8s-app: kube-dns
-      ports:
-        - protocol: UDP
-          port: 53
-        - protocol: TCP
-          port: 53
-
-    # LiteLLM — model inference proxy.
-    - to:
-        - namespaceSelector:
-            matchLabels:
-              kubernetes.io/metadata.name: litellm
-          podSelector:
-            matchLabels:
-              app.kubernetes.io/name: litellm
-      ports:
-        - protocol: TCP
-          port: 4000
-
-    # Brain MCP — knowledge base query/write.
-    - to:
-        - namespaceSelector:
-            matchLabels:
-              kubernetes.io/metadata.name: ingestion
-          podSelector:
-            matchLabels:
-              app.kubernetes.io/name: brain-mcp
-      ports:
-        - protocol: TCP
-          port: 8080
-
-    # Gitea MCP — repo/issue/PR access.
-    - to:
-        - namespaceSelector:
-            matchLabels:
-              kubernetes.io/metadata.name: gitea
-          podSelector:
-            matchLabels:
-              app.kubernetes.io/name: gitea-mcp
-      ports:
-        - protocol: TCP
-          port: 8080
-
-    # OTLP — trace export to Jaeger collector.
-    - to:
-        - namespaceSelector:
-            matchLabels:
-              kubernetes.io/metadata.name: observability
-          podSelector:
-            matchLabels:
-              app.kubernetes.io/name: jaeger
-      ports:
-        - protocol: TCP
-          port: 4318
----
-apiVersion: networking.k8s.io/v1
-kind: NetworkPolicy
-metadata:
-  name: __PROJECT_NAME__-default-deny
-  namespace: __PROJECT_NAME__
-  labels:
-    app.kubernetes.io/name: __PROJECT_NAME__
-spec:
-  podSelector:
-    matchLabels:
-      app.kubernetes.io/name: __PROJECT_NAME__
-  policyTypes:
-    - Ingress
-    - Egress
-  # Empty rules = deny-all. The allow-list above is additive on Egress.
-  # Ingress stays denied unless a sibling policy opens specific ports.
diff --git a/scripts/context-sync.sh b/scripts/context-sync.sh
deleted file mode 100755
index 4f7300e..0000000
--- a/scripts/context-sync.sh
+++ /dev/null
@@ -1,201 +0,0 @@
-#!/usr/bin/env bash
-# Generates harness-specific context files from .context/PROJECT.md
-# Project-level script — run from a project directory.
-#
-# For Claude Code: generates project-only CLAUDE.md (it inherits root via tree walk)
-# For everything else: concatenates root AGENT.md + project PROJECT.md
-#
-# Usage: ./scripts/context-sync.sh [--force] [adapter...]
-# Task:  task context:sync
-#
-# Override root context: ROOT_CONTEXT=~/dev/.context/AGENT.md ./scripts/context-sync.sh
-
-set -euo pipefail
-
-# Parse --force flag and collect adapter names separately
-FORCE=false
-ADAPTERS=()
-for _arg in "$@"; do
-  case "$_arg" in
-    --force) FORCE=true ;;
-    *) ADAPTERS+=("$_arg") ;;
-  esac
-done
-
-PROJECT_FILE=".context/PROJECT.md"
-
-# Walk up to find root .context/AGENT.md
-find_root_context() {
-  local dir
-  dir="$(pwd)"
-  while [ "$dir" != "/" ]; do
-    dir="$(dirname "$dir")"
-    if [ -f "$dir/.context/AGENT.md" ]; then
-      echo "$dir/.context/AGENT.md"
-      return
-    fi
-  done
-  echo ""
-}
-
-ROOT_CONTEXT="${ROOT_CONTEXT:-$(find_root_context)}"
-
-if [ ! -f "$PROJECT_FILE" ]; then
-  echo "Error: $PROJECT_FILE not found. Are you in a project root?"
-  exit 1
-fi
-
-# Pre-flight: reject unfilled {{...}} placeholders unless --force
-if [ "$FORCE" = false ]; then
-  _placeholders=$(grep -n '{{[^}]*}}' "$PROJECT_FILE" 2>/dev/null || true)
-  if [ -n "$_placeholders" ]; then
-    echo "Error: unfilled placeholders in $PROJECT_FILE:" >&2
-    while IFS= read -r _match; do
-      _lineno="${_match%%:*}"
-      _content="${_match#*:}"
-      _token=$(printf '%s' "$_content" | grep -o '{{[^}]*}}' | head -1)
-      echo "  $PROJECT_FILE:$_lineno: unfilled placeholder $_token" >&2
-    done <<< "$_placeholders"
-    echo "" >&2
-    echo "Fill these placeholders, then re-run: task context:sync" >&2
-    echo "To bypass validation: bash scripts/context-sync.sh --force" >&2
-    exit 1
-  fi
-fi
-
-if [ -n "$ROOT_CONTEXT" ] && [ -f "$ROOT_CONTEXT" ]; then
-  echo "  Root context: $ROOT_CONTEXT"
-else
-  echo "  No root AGENT.md found (project context only)"
-fi
-
-# Emit root context + separator
-root_block() {
-  if [ -n "$ROOT_CONTEXT" ] && [ -f "$ROOT_CONTEXT" ]; then
-    cat "$ROOT_CONTEXT"
-    echo ""
-    echo "---"
-    echo ""
-  fi
-}
-
-# ── Claude Code ──────────────────────────────────────────────
-# Claude Code walks up the tree — it finds ~/dev/CLAUDE.md automatically.
-# Project-level CLAUDE.md only needs project-specific context.
-generate_claude() {
-  cat "$PROJECT_FILE" > CLAUDE.md
-  echo "  → CLAUDE.md (project-only; Claude Code inherits root)"
-}
-
-# ── AGENTS.md (Crush, Pi, Antigravity) ──────────────────────
-# These tools read AGENTS.md from cwd but don't walk up.
-# Concatenate root + project.
-generate_agents() {
-  { root_block; cat "$PROJECT_FILE"; } > AGENTS.md
-  echo "  → AGENTS.md (root + project; Crush, Pi, Antigravity)"
-}
-
-# ── Cursor ───────────────────────────────────────────────────
-generate_cursor() {
-  {
-    echo "# Cursor rules — auto-generated"
-    echo "# Do not edit. Run: task context:sync"
-    echo ""
-    root_block
-    cat "$PROJECT_FILE"
-  } > .cursorrules
-  echo "  → .cursorrules (root + project)"
-}
-
-# ── Aider ────────────────────────────────────────────────────
-generate_aider() {
-  { root_block; cat "$PROJECT_FILE"; } > .aider.conventions.md
-  if [ ! -f .aider.conf.yml ]; then
-    cat > .aider.conf.yml << 'YAML'
-read: .aider.conventions.md
-auto-commits: false
-YAML
-  fi
-  echo "  → .aider.conventions.md (root + project)"
-}
-
-# ── Generic system prompt (Open WebUI, Mods, etc.) ──────────
-generate_system_prompt() {
-  {
-    echo "You are a coding assistant working on a specific project."
-    echo "Follow all conventions from both the root agent context and project context."
-    echo ""
-    echo "---"
-    echo ""
-    root_block
-    cat "$PROJECT_FILE"
-    echo ""
-    echo "---"
-  } > .context/system-prompt.txt
-  echo "  → .context/system-prompt.txt (root + project)"
-}
-
-# ── MCP config ───────────────────────────────────────────────
-generate_mcp() {
-  # Ensure baseline file exists with project-specific knowledge server
-  if [ ! -f .context/mcp.json ]; then
-    cat > .context/mcp.json << 'JSON'
-{
-  "mcpServers": {
-    "knowledge": {
-      "url": "http://localhost:3100/mcp",
-      "description": "Project knowledge base — vector + graph retrieval"
-    }
-  }
-}
-JSON
-  fi
-
-  # Merge root mcp-servers.json if found alongside root AGENT.md
-  local root_mcp=""
-  if [ -n "$ROOT_CONTEXT" ] && [ -f "$ROOT_CONTEXT" ]; then
-    local candidate
-    candidate="$(dirname "$ROOT_CONTEXT")/mcp-servers.json"
-    [ -f "$candidate" ] && root_mcp="$candidate"
-  fi
-
-  if [ -z "$root_mcp" ]; then
-    echo "  → .context/mcp.json (exists, no root mcp-servers.json found)"
-    return
-  fi
-
-  # Root servers take precedence over project entries on key conflict
-  local root_servers count updated
-  root_servers=$(jq '.servers' "$root_mcp")
-  count=$(printf '%s' "$root_servers" | jq 'keys | length')
-  updated=$(jq --argjson root "$root_servers" \
-    '.mcpServers = (.mcpServers + $root)' \
-    .context/mcp.json)
-  printf '%s\n' "$updated" > .context/mcp.json
-  echo "  → .context/mcp.json (merged $count root servers)"
-}
-
-echo "Syncing project context from $PROJECT_FILE..."
-
-if [ ${#ADAPTERS[@]} -eq 0 ]; then
-  generate_claude
-  generate_agents
-  generate_cursor
-  generate_aider
-  generate_system_prompt
-  generate_mcp
-else
-  for adapter in "${ADAPTERS[@]}"; do
-    case "$adapter" in
-      claude)  generate_claude ;;
-      agents)  generate_agents ;;
-      cursor)  generate_cursor ;;
-      aider)   generate_aider ;;
-      prompt|system|openwebui|owui|generic) generate_system_prompt ;;
-      mcp)     generate_mcp ;;
-      *) echo "Unknown adapter: $adapter" >&2; exit 1 ;;
-    esac
-  done
-fi
-
-echo "Done."