cad-atlas/AGENTS.md

Agent context — Mathias workspace

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.

0. Pre-task ritual — before ANY implementation (non-negotiable). Run this before writing a single line:

   • Query the brain (brain_query) for the domain + symptom. If the result changes your approach, surface it before acting. 5 seconds beats 5 hours.
   • Load the relevant skill — see trigger table in Engineering Skills below.
   • Write the failing test first. Name the test before the function. If the target is untestable (e.g. main() wiring), extract the logic into a testable function first. No implementation without a red test.
   • State the observable success criterion — what specific behavior, output, or passing test proves this is done?

   TDD is non-negotiable. "Tests pass" is not proof of correctness — only proof the tests ran. Write tests that would catch the bug before writing code that fixes it.

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.

   The PR trigger is RISK, not concurrency. risk/low work commits straight to main with CI as the gate — that is most work and is unchanged. risk/medium and above opens a PR, even solo, even with nobody waiting to review it. This mirrors the tiered merge model in assessor-loop method/transfer-to-a-regulated-organisation.md §4.1, so the estate practises what it sells instead of merging everything on green and claiming otherwise.

   Concurrency is NOT a reason to branch. Measured 2026-08-29 across three real collisions between parallel sessions: two were not branching problems at all — a generated artifact written to the same path by both (branching turns that into a merge conflict on a 33k-line JSON), and uncommitted work in a shared checkout (a per-session worktree fixes it) — and the third was an ordinary non-fast-forward push resolved by rebase in seconds. Use a worktree per session on shared repos and timestamp generated artifacts. Do not reach for a branch.

   Branch+PR flow carries documented costs on solo repos: merging auto-closes gated issues, review comments live on a thread separate from the issue's, and a merged-and-deleted branch cannot be amended (brain: gitea-pr-merge-autoclose-and-review-comment-footguns). Those are the price of the risk/medium+ lane — paid deliberately, not incurred by accident.

   Exception — external contributor or client four-eyes requirement: Use PR flow when a human reviewer outside the project is required. Document the reason in PROJECT.md.

6. Close the loop — every substantive task ends with the same ritual. Shipping the code is not the end of the task; capturing it is. Run this unprompted:

   • Tag + bump SemVer on the change (annotated tag; minor for a feature or new/changed ADR, patch for a fix; docs in the same commit). Check the repo's actual last tag — stated versions in docs drift stale.
   • Push main and the tag (CI is the gate).
   • Persist generalizable learnings to the brain (brain_write, wing/hall) — the reusable patterns and the footguns that would bite anyone again, never project status. See Knowledge base — when to write below.
   • File discovered-but-deferred work as tracker issues on the project's own repo — token-budget gaps, recorded ADR limitations, v2 follow-ups. Don't let "out of scope, recorded" rot in a commit message; make it a ticket with a source pointer.
   • Surface the brain entries and issue numbers in the closing summary so the trail is auditable.

   Chat is not a durable store. Never rely on Mathias remembering anything. Every gap, caveat, risk, open question and next step must land in one of four places before the turn ends: the brain, a repo (code, docs, ADR), a tracker issue, or a plan file. A finding stated only in conversation is lost the moment the session closes — and this estate has repeatedly proved that a remembered convention is an unreliable control.

   This applies hardest to what is tempting to leave as a closing remark: "worth knowing X isn't enforced yet", "this is a convention, not a gate", "we should revisit Y". Those are exactly the items that need a locator. If it is worth saying in a summary, it is worth writing where it can be found again without you.

CAD sprint patterns

Two CAD sprint archetypes (defs: brain wiki/homelab/decisions/continuous-agentic-development-cad-concept-2026-06-16):

• Inception Sprint — birth a new repo. Oath-gated close; tag only once every clause holds. Brain: .../inception-sprint-and-oath.
• Sweep Sprint — clear an existing backlog. Parallel read-only research per issue → sequential one-issue-per-commit execution. Brain: .../sweep-sprint-and-oath.

Shared: decisions that are yours, not the agent's, surface as DECISION NEEDED (question/options/recommendation/blocks) — ask live if you're present, else comment and wait, never act past it. Live-breaking regressions get git revert, never forward-fixed; host-network work skips even read-only research (solo, synchronous, you present). Rule 6's closing ritual only fires once every Oath clause genuinely holds — blocked = descoped-and-tracked, never marked done.

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)	stdlib logging w/ structured extra= (or structlog)	—
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
• Python style (fallback language): ruff (format+lint, one tool), mypy --strict (non-negotiable, matches Go's static typing discipline), pytest + pytest-cov (table-driven via @pytest.mark.parametrize), uv (venv+deps+lock, one tool), pydantic-settings (typed env-var config — same principle as Go's typed structs), src-layout + pyproject.toml only (no setup.py)
• Errors: fmt.Errorf("operation: %w", err) — never naked, never log-and-return. Python: raise X from e (exception chaining, same principle) — never bare except, never silent pass
• 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

Secret handling (every harness, every command)

Tool output is persisted: terminal → ~/.claude/projects transcripts → claudewatcher → brain/wiki → gitea history. A secret printed once is searchable forever, and clearing it means rotating the key. So:

1. Never print, echo, log, or transform a secret to inspect it. No base64/xxd/cat of a key, and never pipe a secret through a transform to defeat op run's output masking (it masks raw values; base64 hides them from the mask — that exact trick leaked a key on 2026-06-11).
2. Secrets stay in the subprocess. Reference them only as env vars consumed inside op run --env-file ~/.op-env -- <cmd>. Never place a literal secret in a command's argv (it lands in the tool call and the transcript).
3. Existence check without revealing the value: [ -n "$X" ] && echo set — never ${X:-...} (returns the value when set) and never echo a substring of it.
4. Cross-host secrets: run the secret-consuming command on the host that has the secret; do not forward a raw key over ssh argv/stdout.
5. If a secret does leak into output, say so immediately and flag it for rotation — don't bury it.

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	Apple Silicon Mac, ~/dev is here

• Model routing: LiteLLM in front of llama-swap (local) + cloud APIs (when permitted)
• Orchestration: k3s on koala only — single control-plane node, no workers (verified 2026-08-30). iguana and flamingo are on the tailnet but are not cluster members
• 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://git.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

Supported harnesses (2026-09-01) — Cursor and Aider were dropped; do not re-add adapters for them:

• Claude Code, Antigravity, OpenCode, Mistral Vibe, Crush, Pi Coding Agent
• Open WebUI, Charmbracelet Mods, or any tool that takes a system prompt string
• Any other tool that reads AGENTS.md, which is the de facto convention

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, .context/system-prompt.txt

Repos still carrying .cursorrules, .aider.conventions.md or .aider.conf.yml are stale: nothing regenerates them. Delete them the next time you touch the repo, along with their entries in that repo's task check assertion.

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. Which file a harness reads does NOT split cleanly by lookup strategy, so do not reason from the strategy:

   • CLAUDE.md — Claude Code (walks up the tree)
   • AGENTS.md — Crush, Pi, Antigravity (cwd-only), OpenCode (project root, global at ~/.config/opencode/AGENTS.md), and Mistral Vibe, which walks up from cwd and takes the first AGENTS.md it finds (user-level ~/.vibe/AGENTS.md)
   • .context/system-prompt.txt — Open WebUI, Mods, raw API calls; not read by any harness, it is for pasting

   Vibe tree-walks for AGENTS.md, so "tree-walkers use CLAUDE.md" is false. Each repo's AGENTS.md embeds the canonical and is the first one found from inside that repo, so this works — but by construction, not by luck.

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 live in the mathias/skills repo (git.d-ma.be/mathias/skills). Load at task start — not "on demand" but on schedule, before writing code. Browse SKILLS_INDEX.md for the full list.

Claude Code installs the repo as a plugin from the marketplace the same repo hosts. That path also carries the secret-guard and pre-task-ritual hooks and the brain/gitea/infra MCP servers, which symlinks cannot:

/plugin marketplace add https://git.d-ma.be/mathias/skills.git
/plugin install skills@mathias

Skills are namespaced under Claude Code: load tdd as skills:tdd. The trigger table below uses bare names — prefix them.

Crush, Pi and OpenCode have no plugin system and are wired by install.sh as one directory symlink per harness via ~/.agents/skills. Do not clone the checkout into ~/dev: that folder is iCloud-shared between the Macs and spans two path roots (/Users/mathias vs /home/mathias on koala), so a checkout there is a git working tree with two concurrent writers, and any absolute symlink stored in it is broken on one side by construction (skills#23). Leave SKILLS_CHECKOUT_DIR at its host-local default.

Check a host with task doctor in the skills repo; it also runs automatically at session start via the plugin.

Skill trigger table — load before starting, not after getting stuck:

Task type	Load
Any feature or bug fix	tdd
Refactor or design	clean-code or solid
Debug	problem-analysis
Review code or PRs	code-review
Frame a problem before coding	problem-analysis

---

cad-atlas

Identity

• Name: cad-atlas
• Owner: Mathias
• Client: personal
• Repo: git.d-ma.be/mathias/cad-atlas
• Status: active
• Stack: Go + Templ + HTMX + CDN Tailwind (template-go-web). Cross-project conventions: ~/dev/.context/AGENT.md.

What this is

A visual atlas of the Continuous Agentic Development (CAD) workflow — the full path from a captured signal to a deployed k3s pod, one screen, reel-style ("From Signal to Pod"). It exists to (a) make the homelab's agentic delivery pipeline legible to a human, and (b) render the CAD audit chain — which doubles as the regulated-industry audit artifact.

Core thesis: the CAD audit chain is the visualization data. TELOS → goal → spec → issue → execution → attestation → deploy is both the trace and the audit package. Phase C renders it once and serves two masters (observability + compliance).

Phases

• Phase A — static hero viz (current). Self-contained internal/web/static/cad-atlas.html, data-driven from a hand-authored STAGES array (ground-truth snapshot from brain, 2026-07-19). Served at / by internal/web/handler.go via go:embed. Reel-parity: SVG spine with arrowheads, animated pulse, dashed feedback bus (stage 08 → TELOS), replay + slow-mo.
• Phase B — generated-from-source. Parse brain docs + .gitea/workflows + infra manifests → render the graph so it can't drift from config.
• Phase C — live trace viewer. Replace the static STAGES array with live reads of the assessor-loop attestation ledger + brain session_log + Gitea run API + Flux events. This is the prize: a real signal→pod trace viewer that is also the audit package.

The workflow it visualizes (9 stages)

00 Signals (Applied AI Radar → mathias/signals) → 01 TELOS (intention substrate) → 02 Strategic session (claude.ai frontier + LLM Council + Autoresearch Council) → 03 Spec → Gitea issue (agent-ready contract; Ed25519 admission #36; var-go Oath) → 04 Human dispatch gate (the only checkpoint; session-dispatch bridge → cad-dispatch.yml) → 05 Execute · agentsquad (serve/taskqueue, exec+review loop, risk LOW/MED/HIGH, dma-cli routing, assessor-loop ledger) → 06 PR → CI (go test/vet/lint/govulncheck + var-go/oath gate) → 07 CD → pod (Flux GitOps → k3s on koala) → 08 Loop back (outcome scored vs TELOS goal).

Three orthogonal governance gates

Gate	Guards	Where
Ed25519 admission controller (#36)	spec integrity (issue untampered)	stage 03
dispatch-allow (.dispatch-allow + mathias/dispatch allowlist)	repo eligibility (may agents run here)	stage 04/05
var-go Oath (cmd/vargo-gate, commit status var-go/oath)	output correctness (PR satisfies the Oath; floor over reviewer, anti-rubber-stamp #55)	stage 06

Dogfooding

This repo is built through the workflow it depicts. It is dispatch-allow-enabled, and its own build increments are governed by a var-go Oath embedded in their spec issues (see the Stage-03 tracking issue). cmd/vargo-gate is wired into .gitea/workflows/cd.yml's oath job — on every pull_request it fetches the linked issue's oath and gates cad-atlas's own real candidate (oathcandidate/, #8): it parses the committed .gitea/workflows/cd.yml and checks the oath job actually exists and invokes cmd/vargo-gate, then posts the var-go/oath commit status. This is a real check (TDD'd: passes on the real file, fails closed on a fixture missing the job), not swedsl's toy self-test stub — swedsl#35 (import path) and swedsl#38 (real-candidate subprocess gating) unblocked it. Required by branch protection on main (#8, closed 2026-07-20), confirmed holding on a real PR. Direct pushes remain allowlisted for mathias per this repo's TBD convention. Disclosed in the CI config comment, this doc, and docs/INCEPTION-OATH.md.

Brain references (source of truth — brain_get <path>)

• wiki/homelab/decisions/cad-atlas-audit-chain-is-viz-data.md — this project's genesis note: the audit-chain-is-viz-data thesis, the three gates, template-go-web footguns
• wiki/homelab/decisions/inception-sprint-and-oath.md — the Inception Sprint + Oath methodology (this repo's birth is the worked example); see docs/INCEPTION-OATH.md for the kept Oath
• knowledge/workflow-idea-to-running-service.md — Double Diamond idea→service workflow
• wiki/homelab/decisions/continuous-agentic-development-cad-concept-2026-06-16.md — CAD definition
• wiki/agentsquad/decisions/cad-dispatch-bridge.md — claude.ai → agentsquad trigger path
• wiki/agentsquad/facts/llm-council-design-and-first-runs-2026-06-21.md — LLM Council
• wiki/agentsquad/decisions/autoresearch-council-sibling-pipe.md — Autoresearch Council
• wiki/agentsquad/decisions/serve-http-task-api.md — agentsquad serve/taskqueue
• knowledge/var-go-anchor-to-span-spike-verdict.md — var-go runner + CAD gate seam
• knowledge/swedsl-vargo-sprint1-enforcement-teeth-verdict.md — vargo-gate enforcement teeth
• wiki/assessor-loop/decisions/assessor-loop-genesis.md — attestation ledger (Phase C source)
• wiki/homelab/facts/homelab-network-topology-reference.md — koala/iguana/flamingo/piguard

External references

• Inspiration reel — "From Inbox to Shipped" pipeline viz: https://www.instagram.com/reel/DY92L7bu27j/
• karpathy/llm-council — origin of the Council pattern
• Double Diamond design process (Discover/Define/Develop/Deliver)

Run

task check      # lint + vet + test (CI gate)
task run        # build + serve at http://localhost:8080  → the atlas

Deploy note

CD (.gitea/workflows/cd.yml) deploys to k3s namespace cad-atlas via Flux. Per the known template-go-agent CD gap: the deploy job stays RED until mathias/infra has k3s/apps/cad-atlas/deployment.yaml. check + build are the real bootstrap gate.