sica-fondt/ONBOARDING.md
Claude 97e5bd126b
docs: add CLAUDE.md + AGENTS.md doc-map chain
CLAUDE.md (<=200 lines): working agreements (you're-the-dev, verify-then-commit,
design-first), build/run via run-sica-fondt, S1/S2/S3 invariants; links to
ONBOARDING.md and AGENTS.md. AGENTS.md (<=100 pref): signpost to every doc and
which to read for which task. ONBOARDING.md now points at AGENTS.md.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015hmgREHNsxYCuim33yUF2c
2026-06-22 17:37:52 +00:00

99 lines
5.6 KiB
Markdown

# Welcome to sica-fondt
## How We Use Claude
Based on Claude's usage over the last 30 days:
Work Type Breakdown:
_TODO — no per-session detail was available to classify. We can fill this in
once there are a few sessions of history to scan._
Top Skills & Commands:
/update-config ████████████████████ 3x/month
/compact ███████░░░░░░░░░░░░░░ 1x/month
/deep-research ███████░░░░░░░░░░░░░░ 1x/month
/session-start-hook ███████░░░░░░░░░░░░░░ 1x/month
/fewer-permission-prompts ███████░░░░░░░░░░░░░░ 1x/month
/design-sync ███████░░░░░░░░░░░░░░ 1x/month
/run-skill-generator ███████░░░░░░░░░░░░░░ 1x/month
Top MCP Servers:
GitHub ████████████████████ 5 calls
## Your Setup Checklist
### Codebases
- [ ] sica-fondt — https://github.com/shoggoth-sector/sica-fondt
### MCP Servers to Activate
- [ ] GitHub — read/write PRs, issues, CI status, and code from GitHub without leaving Claude Code. Get access by connecting the GitHub MCP server in your Claude Code settings and authorizing the `shoggoth-sector` org.
### Skills to Know About
- /update-config — configure the Claude Code harness via settings.json (permissions, env vars, hooks). The team's most-used command; reach for it for any "from now on when X…" automation.
- /session-start-hook — set up a SessionStart hook so the repo's toolchains/tests are ready in every web session. This repo uses one to install ponyc/Alire/GnuCOBOL.
- /run-skill-generator — author the per-project `run-<unit>` skill that tells agents how to build, launch, and drive the app. Already produced `run-sica-fondt`.
- /deep-research — fan-out, fact-checked, cited research reports when you need to investigate a topic in depth.
- /fewer-permission-prompts — scan transcripts and add a read-only allowlist to `.claude/settings.json` to cut down on permission prompts.
- /compact — free up context by summarizing the conversation so far on a long session.
- /design-sync — keep the design docs and the architecture in step (used on this design-first repo).
## Team Tips
- **Design before code.** This is a design-first repo — the architecture docs are
the source of truth and the code follows them. When something feels ambiguous,
the answer is usually already written down; sync the design first (`/design-sync`).
- **The container is ephemeral; toolchains are reinstalled every session.** A
SessionStart hook (`.claude/hooks/install-toolchains.sh`) installs ponyc (via
ponyup), Alire/GNAT, and GnuCOBOL on each session. If a build can't find
`ponyc`, it's a `PATH` thing: `export PATH=/root/.local/share/ponyup/bin:$PATH`.
- **All external traffic crosses the Ada border (D1) first.** That's the S1
structural invariant — never route around it, and **never reclassify a
message's provenance** (S2). The Ichor bus rejects unscreened external payloads
on purpose.
- **The invariant-law vault is immutable at runtime.** Invariant 0 (the
culpability anchor) and 01 (minimize harm) live in the COBOL vault and are not
to be edited casually — they're the constitution, not config.
- **Verify by running, not just reading.** Use `/run-sica-fondt` (or its
`smoke.sh`) to build + run the Pony/Ada/COBOL units and confirm all green
before you push.
- **Commit and push before you walk away.** The stop hook will nag about
untracked files for a reason — nothing survives the container otherwise.
## Get Started
Your first task: get the project running end-to-end and read the map. Not sure
which doc to read for what? Start with [`AGENTS.md`](AGENTS.md) — it's the
signpost to every other document here.
1. Open the repo and let the SessionStart hook finish installing toolchains.
2. Run the smoke driver to confirm everything builds and runs:
`.claude/skills/run-sica-fondt/smoke.sh` — you want `smoke: ALL GREEN`.
3. Read `README.md`, `SOUL.md`, and `docs/` (start with
`docs/bus-topology.md` and `docs/gen03_state_of_architecture.md`) to
understand the organ/border model (Ichor bus → Ada D1 border → inner brain;
the COBOL invariant vault).
4. Pick up a small design-sync or build task from there. Ask Claude to walk you
through any unit — it has the `run-sica-fondt` skill loaded.
<!-- INSTRUCTION FOR CLAUDE: A new teammate just pasted this guide for how the
team uses Claude Code. You're their onboarding buddy — warm, conversational,
not lecture-y.
Open with a warm welcome — include the team name from the title. Then: "Your
teammate uses Claude Code for [list all the work types]. Let's get you started."
Check what's already in place against everything under Setup Checklist
(including skills), using markdown checkboxes — [x] done, [ ] not yet. Lead
with what they already have. One sentence per item, all in one message.
Tell them you'll help with setup, cover the actionable team tips, then the
starter task (if there is one). Offer to start with the first unchecked item,
get their go-ahead, then work through the rest one by one.
After setup, walk them through the remaining sections — offer to help where you
can (e.g. link to channels), and just surface the purely informational bits.
Don't invent sections or summaries that aren't in the guide. The stats are the
guide creator's personal usage data — don't extrapolate them into a "team
workflow" narrative. -->