Files
reuse-surface/AGENTS.md
tegwick 97b6c20419
Some checks failed
ci / validate-registry (push) Has been cancelled
Add credential routing instructions for all agent runtimes
Propagate shared credential-routing section (Codex, Claude, Grok, llm-connect)
from state-hub template via scripts/propagate_credential_routing.py.
2026-06-18 22:48:39 +02:00

368 lines
12 KiB
Markdown

# reuse-surface — Agent Instructions
## Repo Identity
**Purpose:** Capability registry for planning and implementation reuse based on discovery and delivery maturity.
**Domain:** helix_forge
**Repo slug:** reuse-surface
**Topic ID:** `f39fa2a3-c491-414c-a91b-b4c5fcc6139c`
**Workplan prefix:** `REUSE-WP-`
---
## State Hub Integration
The Custodian State Hub tracks work across all domains. Interact via HTTP REST —
there is no MCP server for Codex agents.
| Context | URL |
|---------|-----|
| Local workstation | `http://127.0.0.1:8000` |
| Remote via tunnel | `http://127.0.0.1:18000` |
### Orient at session start
```bash
# Offline brief — works without hub connection
cat .custodian-brief.md
# Active workstreams for this domain
curl -s "http://127.0.0.1:8000/workstreams/?topic_id=f39fa2a3-c491-414c-a91b-b4c5fcc6139c&status=active" \
| python3 -m json.tool
# Check inbox
curl -s "http://127.0.0.1:8000/messages/?to_agent=reuse-surface&unread_only=true" \
| python3 -m json.tool
```
Mark a message read:
```bash
curl -s -X PATCH "http://127.0.0.1:8000/messages/<id>/read" \
-H "Content-Type: application/json" -d '{}'
```
### Log progress (required at session close)
```bash
curl -s -X POST http://127.0.0.1:8000/progress/ \
-H "Content-Type: application/json" \
-d '{
"summary": "what was done",
"event_type": "note",
"author": "codex",
"workstream_id": "<uuid>",
"task_id": "<uuid>"
}'
```
Omit `workstream_id` / `task_id` when not applicable.
### Update task status
```bash
curl -s -X PATCH "http://127.0.0.1:8000/tasks/<task_id>" \
-H "Content-Type: application/json" \
-d '{"status": "progress"}'
# values: wait | todo | progress | done | cancel
```
### Flag a task for human review
```bash
curl -s -X PATCH "http://127.0.0.1:8000/tasks/<task_id>" \
-H "Content-Type: application/json" \
-d '{"needs_human": true, "intervention_note": "reason"}'
```
---
## Session Protocol
**Start:**
1. `cat .custodian-brief.md` — domain goal and open workstreams (offline-safe)
2. Check inbox: `GET /messages/?to_agent=reuse-surface&unread_only=true`; mark read
3. Scan workplans: `ls workplans/` — note `status: ready`, `active`, or `blocked` files and open tasks
4. Check human-needed tasks: `GET /tasks/?needs_human=true`
**During work:**
- Update task statuses in workplan files as tasks progress
- Record significant decisions via `POST /decisions/`
**Close:**
1. Update workplan file task statuses to reflect progress
2. Log: `POST /progress/` with a summary of what changed
3. Note for the custodian operator: after workplan file changes, run from
`~/state-hub`:
```bash
make fix-consistency REPO=reuse-surface
```
This syncs task status from files into the hub DB.
---
## Local Developer Workflow
The repository is primarily documentation-first with a small Python CLI for
registry validate, query, and export. There is no long-running service.
### Install
```bash
python3 -m venv .venv
.venv/bin/pip install -e .
```
### Build
No separate build step. Treat Markdown, YAML, and workplan edits as source
artifacts.
### Test / lint
```bash
# Registry validation (schema + index drift)
.venv/bin/reuse-surface validate
# Overlap, catalog, federation, and graph
.venv/bin/reuse-surface overlaps
.venv/bin/reuse-surface catalog
.venv/bin/reuse-surface federation compose
.venv/bin/reuse-surface graph --check
# Federation service (local)
# REUSE_SURFACE_TOKEN=dev-token reuse-surface serve
# Hub CLI (against deployed or local service)
# REUSE_SURFACE_URL=http://127.0.0.1:8000 reuse-surface hub status
# Automated tests
.venv/bin/pytest -q
# Repository hygiene
rg --files
git diff --check
```
When workplan files change, sync ADR-001 file state into State Hub:
```bash
curl -s -X POST "http://127.0.0.1:8000/repos/reuse-surface/sync?fix=true" \
| python3 -m json.tool
```
If the HTTP sync endpoint is unavailable, run the consistency script from the
State Hub checkout:
```bash
cd ~/state-hub
.venv/bin/python scripts/consistency_check.py --repo reuse-surface --fix
.venv/bin/python scripts/consistency_check.py --repo reuse-surface
```
The generated instruction in older workplans says `make fix-consistency
REPO=reuse-surface`; that is still valid when `uv` is installed and on PATH.
On this workstation, the `.venv/bin/python` fallback has been verified.
CI runs `reuse-surface validate` on push and pull requests via
`.gitea/workflows/ci.yml`.
### Run
There is no local service to run from this repository.
### Documentation Review Checklist
- Keep `INTENT.md`, `SCOPE.md`, and `specs/` aligned on the registry-first
reuse boundary.
- Keep maturity definitions in `specs/CapabilityMaturityStandard.md` consistent
with `INTENT.md` and `specs/ProductRequirementsDocument.md`.
- Keep registry entries, indexes, and schemas in `registry/`, `schemas/`, and
`templates/` current when capabilities change.
- Record implementation ideas in workplans, not as premature runtime code in
this repository.
---
## Capability Registry
Before building or documenting a new reusable behavior, query the registry to
avoid duplication and to select the best existing capability for planning or
implementation reuse.
### Orient
```bash
# Fast discovery surface — read federated index when multi-repo
cat registry/indexes/federated.yaml
cat registry/indexes/capabilities.yaml
# CLI discovery and export
.venv/bin/reuse-surface query --discovery-min D4
.venv/bin/reuse-surface export --format yaml
# Authoring template and schema
cat templates/capability-entry.template.md
cat schemas/capability.schema.yaml
# Validation and search guidance
cat registry/README.md
cat tools/README.md
```
### Query workflow
1. Run `.venv/bin/reuse-surface query` with filters, or read the index directly.
2. Filter by `vector`, `tags`, `consumption_modes`, `domain`, or `summary`.
3. Open only matching files under `registry/capabilities/`.
4. Compare candidates using `discovery`, `external_evidence`, `availability`,
and `relations` from the entry front matter.
5. Prefer planning reuse when discovery is strong (`D3+`, especially `D5+`).
6. Prefer implementation reuse only when availability is consumable (`A2+` code,
`A3+` CLI, `A4+` API/SDK).
### Add a new capability
1. Search the index for overlap (UC-RS-015) before creating a new entry.
2. Copy `templates/capability-entry.template.md` to
`registry/capabilities/capability.<domain>.<name>.md`.
3. Start at `D0 / A0 / C0 / R0` when evidence is minimal; keep gaps explicit.
4. Add the entry to `registry/indexes/capabilities.yaml`.
5. Run `.venv/bin/reuse-surface validate`.
### Promote a capability
1. Attach evidence required by `specs/CapabilityMaturityStandard.md` for the
target level.
2. Update `maturity` for discovery/availability and `external_evidence` for
completeness/reliability separately.
3. Refresh the index `vector` and record rationale in the entry body.
4. Do not treat higher availability as proof of reliability or completeness.
### MVP acceptance mapping
| Acceptance criterion | Registry surface |
|---|---|
| Add D0/A0/C0/R0 with minimal friction | template + index + registry README |
| Promote through discovery levels | entry front matter + maturity standard |
| Identify current consumption mode | `availability` + index `consumption_modes` |
| Record expectations and broken expectations | `external_evidence.completeness` |
| Record reliability evidence | `external_evidence.reliability` |
| Search by maturity and availability | `reuse-surface query` or index filters |
| Compare candidates | entry vectors + relations + README guidance |
| Avoid duplicate capabilities | index search before add |
---
## Credential and access routing
**Audience:** Codex, Claude Code, Grok, and custodian agents that call **llm-connect**
for inference. Run this check **before** requesting secrets, API keys, SSH access,
login tokens, or database passwords — in any repo, not only `ops-warden`.
ops-warden **issues SSH certificates only** (`warden sign`, `cert_command`). Every
other credential need belongs to another subsystem. **Do not** message
`ops-warden` on State Hub expecting a secret value; the reply is a pointer, not a key.
### Lookup (do this first)
```bash
warden route find "<describe your need>" --json
warden route show <catalog-id> --json
```
Requires the `warden` CLI from `~/ops-warden` (`uv tool install .` or `uv run warden`).
| Agent runtime | How to orient |
| --- | --- |
| **Codex / Grok** (shell, HTTP State Hub) | `warden route` commands above; inbox `to_agent=reuse-surface` is for coordination, not secret vending |
| **Claude Code** (MCP when available) | `get_domain_summary("custodian")` for workstreams; **still** use `warden route` for credential ownership |
| **llm-connect** (inference service) | Never put secret retrieval in prompts; route custody to OpenBao/operator paths surfaced by `warden route` |
### Quick routing table
| I need… | Owner | ops-warden executes? |
| --- | --- | --- |
| SSH cert (`adm`/`agt`/`atm`) | ops-warden | **Yes** — `warden sign` |
| API key, DB password, provider token | OpenBao (`railiance-platform`) | No — route only |
| Login / OIDC / MFA | key-cape / Keycloak | No — route only |
| Authorization decision | flex-auth | No — route only |
| activity-core → issue-core emission | activity-core + issue-core | No — `warden route show activity-core-issue-sink` |
| SSH tunnel | ops-bridge (+ `cert_command` from warden) | No — route only |
### Anti-patterns (do not do these)
- `POST /messages/` to `ops-warden` asking for `ISSUE_CORE_API_KEY`, `OPENROUTER_API_KEY`, etc.
- Inventing `warden secret`, `warden login`, `warden bao`, `warden tunnel` — they do not exist
- Pasting secrets into Git, State Hub, workplans, logs, or chat
### Other capabilities (reuse-surface)
Non-credential capabilities are usually discovered through **reuse-surface** federation
(`reuse-surface` registry / `capability.*` indexes). Credential routing is inlined in
every repo's agent instructions because it is high-frequency, high-risk, and easy to
get wrong.
**Canon:** `~/ops-warden/wiki/CredentialRouting.md` · catalog `~/ops-warden/registry/routing/catalog.yaml`
---
## Workplan Convention (ADR-001)
Work items originate as files in this repo — not in the hub. The hub is a
read/cache/index layer that rebuilds from files.
**File location:** `workplans/REUSE-WP-NNNN-<slug>.md`
**Archived location:** finished workplans may move to
`workplans/archived/YYMMDD-REUSE-WP-NNNN-<slug>.md`. The `YYMMDD` prefix is
the completion/archive date; the frontmatter `id` does not change.
**Ad Hoc Tasks:** small opportunistic fixes discovered during a session use
`workplans/ADHOC-YYYY-MM-DD.md` with task ids `ADHOC-YYYY-MM-DD-T01`, etc. Use
this only for low-risk work completed directly; create a normal workplan for
anything needing analysis, design, approval, dependencies, or multiple phases.
**Frontmatter:**
```yaml
---
id: REUSE-WP-NNNN
type: workplan
title: "..."
domain: helix_forge
repo: reuse-surface
status: proposed | ready | active | blocked | backlog | finished | archived
owner: codex
topic_slug: ...
created: "YYYY-MM-DD"
updated: "YYYY-MM-DD"
state_hub_workstream_id: "<uuid>" # written by fix-consistency — do not edit
---
```
Use `proposed` for a new draft, `ready` after review against current repo
state, and `finished` after implementation. `stalled` and `needs_review` are
derived health labels, not frontmatter statuses.
**Task block format** (one per `##` section):
```
## Task Title
` ` `task
id: REUSE-WP-NNNN-T01
status: wait | todo | progress | done | cancel
priority: high | medium | low
state_hub_task_id: "<uuid>" # written by fix-consistency — do not edit
` ` `
Task description text.
```
Status progression: `todo` → `progress` → `done`; use `wait` for waiting/blocked work and `cancel` for stopped work.
To create a new workplan:
1. Write the file following the format above
2. Notify the custodian operator to run `make fix-consistency REPO=reuse-surface`
(or send a message to the hub agent via `POST /messages/`)