# pi-agent-config

Authoritative configuration for every [Pi](https://pi.dev) agent on this host:
the user-level Pi runtime, plus one directory per dedicated agent scenario.

The live locations (`~/.pi/agent`, `~/pi-workspaces/*`) are **runtime copies**.
Changes are made here and pushed out with `scripts/deploy-*.sh`, never the other
way round — except for scenarios explicitly marked `deploy = "mirror"`, which
record what a running service already does.

Verified against **pi 0.84.3**. Re-run the probes after every `pi update`.

## Gitea repository and push authentication

Everything here — the runtime config, every scenario's agent config, and (for
scenarios that run a service) the application backend — lives in a single Gitea
repository: **`kai/pi-agent-config`** on the server at `192.168.50.45` (web UI on
`:3000`, SSH on `:222`). There is no per-agent repository; a Pi agent is a
directory under `scenarios/`, not a separate repo.

Push and pull over **SSH key auth only** (no HTTP token). The Git user is `git`,
not `kai`. `~/.ssh/config`:

```bash
Host gitea-45
    HostName 192.168.50.45
    User git
    Port 222
    IdentityFile ~/.ssh/id_ed25519_gitea
    IdentitiesOnly yes
```

The remote is `origin -> gitea-45:kai/pi-agent-config.git` (branch `main`). Run
Git from the repository root, never from a scenario subdirectory that might carry
a different remote:

```bash
git pull --rebase origin main
git push origin main
```

Verify access:

```bash
ssh -T gitea-45                             # "Hi there, kai! ... authenticated"
git ls-remote gitea-45:kai/pi-agent-config.git
```

- The key file must be `0600`; never read or print its contents.
- `gitea-pve`, if present in your SSH config, is a host login — not a Git remote.
- The `verify-no-secrets.sh` pre-commit hook must pass; never use `--no-verify`.

## Scenarios

| Scenario | Purpose | Service | Status |
|---|---|---|---|
| [`curator`](scenarios/curator/) | Book / film / TV / music curation agent | `curator.service` | **deployed**; agent config and application backend both tracked here |
| [`memo-inbox`](scenarios/memo-inbox/) | Routes Telegram/WeChat messages to Calendar, Obsidian todo or journal | `pi-memo-telegram.service` | **mirror** of live config, zero behaviour change |
| [`pi-grok`](scenarios/pi-grok/) | Interactive Grok 4.6 coding agent | none (manual) | registered only |

## Application code

A scenario that runs its own service keeps that service's source under
`scenarios/<name>/backend/` — for `curator`, the Python package, tests, systemd
units and config templates. It is the single source of truth; the live checkout
at the scenario's `workspace` path is a runtime copy.

`deploy-scenario.sh` installs only `workspace/` (the `.pi` config) and renders the
launch contract; it does not touch `backend/`, build a venv or restart a service.
Rolling application code out to the live checkout and restarting the unit stays an
operator step, deliberately: restarting decides when to interrupt a live
conversation.

## Start here

| Document | Contents |
|---|---|
| [`docs/pi-runtime-notes.md`](docs/pi-runtime-notes.md) | pi 0.84.3 mechanics that the official docs understate, each verified against source or a live probe |
| [`docs/isolation-baseline.md`](docs/isolation-baseline.md) | The four isolation layers and the flag set every scenario should use |
| [`docs/personality-layering.md`](docs/personality-layering.md) | How to choose between `SYSTEM.md`, `APPEND_SYSTEM.md`, `AGENTS.md`, `SKILL.md` and the request |
| [`docs/gateway-patterns.md`](docs/gateway-patterns.md) | Ten patterns for hosting an agent behind a long-running service |
| [`docs/plans/`](docs/plans/) | Active work plans |

## The finding that motivated this repository

pi emits the `<available_skills>` block **only when a tool named `read` is
active** (`dist/core/system-prompt.js:59,113`). The Curator service ran with
`--no-tools --skill …` for its entire lifetime, so:

- its 64-line media policy never reached the model,
- `--skill` was a no-op,
- and both the README and the deployment runbook described a mechanism that was
  not happening.

Measured on the real workspace, before and after
([evidence](docs/evidence/2026-08-27-curator-phase0-prompt.md)):

| | before | after |
|---|---|---|
| system prompt | 2548 chars | 3539 chars |
| `expert coding assistant` framing | present | **gone** |
| pointer to pi's own documentation | present | **gone** |
| the 64-line media policy | **absent** | present |
| `<available_skills>` | absent | absent |
| workspace `AGENTS.md` | loaded | blocked |
| parent-directory `AGENTS.md` | **leaked** | blocked |

The prompt got *larger*, and that is the fix rather than its cost: of the
original 2548 characters roughly 1.9 KB was pi's coding-assistant scaffolding and
a pointer telling the agent to read pi's documentation and follow its
cross-references — noise for a media agent, and a ready-made escalation path for
injected text — while none of it was Curator's own policy. All 3539 characters
now are.

A mechanism demonstration with a stub `SYSTEM.md` is in
[`docs/evidence/2026-08-27-isolation-probe.md`](docs/evidence/2026-08-27-isolation-probe.md);
its 960-character figure measures the stub, not Curator.

Reproduce either with [`docs/evidence/probe-harness/`](docs/evidence/probe-harness/)
— zero model tokens, because a single RPC `get_state` starts the agent, fires
`session_start` and exits without contacting the provider.

## Layout

```
docs/            mechanics, baselines, plans, reproducible evidence
runtime/agent/   ~/.pi/agent — settings, models.json.template, extensions, prompts
shared/          cross-scenario code
  lib/py/pi_rpc.py            long-lived RPC client with rotation and budgets
  extensions/pi-guard-base.ts path containment, restricted read, capability guard
scenarios/<name>/
  profile.toml   single source of truth for the launch contract
  workspace/     what gets installed into the live workspace
  eval/          recorded golden transcripts
  backend/       application/service code, when the scenario runs its own service
scripts/         diff, deploy, backup, restore, secret guard
secrets/         host-local, untracked
```

## Usage

```bash
# What differs between this repository and the host?
scripts/pi-diff.sh
scripts/pi-diff.sh curator

# Push configuration out (dry run first; neither restarts a service)
scripts/deploy-runtime.sh
scripts/deploy-runtime.sh --apply
scripts/deploy-scenario.sh curator --apply

# Back up the live installation (archives land outside this repository)
scripts/pi-backup.sh
scripts/pi-restore.sh --from <backup dir>

# Verify the mechanics still hold after a pi upgrade
docs/evidence/probe-harness/collect-evidence.sh
shared/extensions/tests/run-guard-checks.sh
python3 shared/lib/py/tests/test_pi_rpc_smoke.py
```

## Secrets

Nothing real ever enters this repository. `models.json`, `auth.json`,
`trust.json` and every `*.env` are ignored; `secrets/` accepts only `.gitkeep`,
`README.md`, `*.example` and `*.template`.

`scripts/verify-no-secrets.sh` enforces this as a pre-commit hook.

Because a scenario's `backend/` now holds real source, the credential-assignment
heuristic requires the value to carry entropy (a digit or uppercase letter):
snake_case identifiers such as `token=extraction_token` are source, not secrets,
while base64/hex/random keys still trip it.

Install it in a fresh clone:

```bash
ln -sf ../../scripts/verify-no-secrets.sh .git/hooks/pre-commit
scripts/verify-no-secrets.sh --all
```

Backups contain plaintext credentials and are written outside the work tree;
`pi-backup.sh` refuses any destination that resolves inside it.

## First-time setup on a new host

```bash
git clone gitea-45:kai/pi-agent-config.git && cd pi-agent-config
ln -sf ../../scripts/verify-no-secrets.sh .git/hooks/pre-commit

cp secrets/zenmux.env.example secrets/zenmux.env
chmod 600 secrets/zenmux.env && $EDITOR secrets/zenmux.env

npm install -g @earendil-works/pi-coding-agent@0.84.3
scripts/deploy-runtime.sh --apply
scripts/deploy-scenario.sh <scenario> --apply
```