Files
vardkey/ai-note.md
T

13 KiB
Raw Blame History

AI Agent Note — Helsesjekk

Short working handover for the next model. Keep it short.

  • TODO.md (Norwegian) is the requirement specification: sections 1–18 are the spec, sections 19–22 a short appendix with the four decisions taken along the way, section 23 the configurable key setup (implemented), and section 24 agreed work that is not implemented yet and comes before steps 9–12.
  • The binding rules live in TODO.md and README.md. This note does not repeat them — it carries the handover, the rules that are easy to get wrong, and the traps that already cost time.
  • HANDOVER.md was deleted as outdated. Do not recreate it.

Status

  • Done and verified: sections 19–22, section 18 steps 1–8, and the whole of section 23 (23.1–23.5). Ticked in TODO.md.
  • Verified how: the automated suite only. The browser flows and the migration of a real pre-section-23 database have not been run by hand yet; they are the manual tests in section 16 of TODO.md.
  • Tests: 178, all green. The user runs them: python3 -m unittest discover -s tests. They run on a Windows copy at C:\Users\Driftsleiar\Downloads\helsesjekk, so they must pass there as well as on Linux.
  • Next work, not started: section 24 — required choices in the form, and the full removal of BIP327/SLIP39. Nothing in section 24 is implemented.
  • After that: section 18 steps 9–12 — the StartOS package, the persistent volume, backup/restore, and .s9pk. Covered by sections 13, 14, 16 and 17 of TODO.md.
  • Future work, deliberately not built: more than three keys, another threshold than 2-of-3, and a user-defined wallet-software list (section 23.6).

Section 24 — agreed, not implemented

Read section 24 of TODO.md for the full list. The parts that are easy to get wrong:

  • Hardware works becomes a required choice between Yes and No, with Select status as a non-selectable first option. It applies to an available hardware key only, and the required flag must be cleared when the key is unavailable, or the browser refuses to submit.
  • Contact confirmation becomes a required choice between Confirmed current and Needs external update, with the same placeholder. It is asked even when the key is unavailable, so it is required there too.
  • NOT_ASSESSED survives only as the derived software_compatibility_status of an unavailable key. No other field offers or accepts it.
  • BIP327/SLIP39 is removed completely: the form section, the validation, the alerts, the history view, the CSV columns, and the tool_checks table with its data. The app is not released and the database is a test database, so those rows may be dropped.
  • The form is renumbered: 01 Schedule, 02 Key, 03 Compatibility, 04 External research, 05 Contact confirmation, 06 Notes. 04 / Standards disappears and the three sections after it shift down one number. 04 / External research is the current security status section, unchanged in content.
  • Schema version 9. The migration drops tool_checks. It is a child table, so no other table is rebuilt.

Section 23.2–23.5 — implemented and verified

Sections 23.2 and 23.3 were delivered together on purpose: the type is free, so the checks must follow the type, and a half-done step would handle a cloud key in the first slot wrongly. Section 23.4 added schema version 8 and its migration, and 23.5 the tests.

  • The setup is stored once. POST /api/setup writes the three key rows; a second call is a 409, because the names and the types are locked (section 23.2). GET /api/setup answers {"configured": …, "keys": […]}.
  • A new database shows the setup step. /setup is a page in STATIC_FILES, and both static/app.js and static/health-check.js send the browser there when /api/setup reports configured: false.
  • An existing database is migrated, not replaced. list_keys() reads the stored rows; seed_fixed_keys() and the fixed device column are gone. A database with 1 or 2 key rows raises RuntimeError and is never answered with the setup step.
  • The name is the product label. ai_security_checks.provider holds the configured key name and the selected wallet software, so its old CHECK list is gone. validation.py builds the list from the key, and static/health-check.js does the same in expectedLabels(). (tool_checks.product worked the same way, but section 24 removes that table.)
  • The type decides, never the number. _key_check() and _compatibility_status() switch on key["type"], and derivedValues() in the script does too.
  • The wallet software choices are Sparrow Wallet plus every configured cloud key name. _wallet_software_choices() builds the list on the server and walletSoftwareChoices() in the script. A hardware key name is never a choice. A cloud key records its own name, because the provider is the software for that key. Section 23.3 writes the choice as Sparrow, but the stored value stayed Sparrow Wallet, so no data migration was needed.
  • The section 20 carry-forward is gone. A submission records the selected key and the selected software only, so carriedSecurity and the "Carried forward from the last saved check" summary are removed from the script.
  • The setup is validated. validate_key_setup() rejects a wrong number of keys, a repeated or out-of-range key number, an empty or over-long name, a name that repeats another one without regard to case, and an unknown type. Name uniqueness is enforced there and not by a UNIQUE constraint, so a legacy database with odd rows is refused by the API instead of being impossible to open.

Known leftovers

  • /api/previous-values still returns a security map that nothing consumes any more. It is deliberate: removing it would shrink the endpoint, but nothing breaks while it stays.

Rules that are easy to get wrong — do not re-open

  • One submission records one key. There is no "round" grouping, and each key's next check is its own date plus one calendar year, computed on the server and not editable in the form.
  • An unavailable key records nothing else. When available is 0, wallet_software and ai_security_checks must be empty; a non-empty value is a 400, not something dropped silently. Nothing is stored for them, and software_compatibility_status becomes NOT_ASSESSED (only NOT_ASSESSED, "" or null may accompany it).
  • Availability is required and offers only Yes/No. There is no "not assessed" value for it, and the server rejects a payload without it.
  • A cloud key collects only Availability. Cloud health, "provider operational" and software compatibility are all derived from it, and its own name is the wallet software.
  • A hardware key needs an explicit compatibility status. OK or PROBLEM; a missing value and NOT_ASSESSED are both a 400 (section 23.1).
  • Hardware works needs an explicit Yes or No for an available hardware key (section 24.1). Until section 24 is implemented, the field may still be left empty.
  • Address path is only recorded when Address path verified is OK. The form enforces this; the server does not.
  • /api/alerts covers every saved check, not only the newest, and merges identical alerts by code plus source. A security alert's source is the provider, which is the key's configured name or the software name.
  • The export is plain CSV only. XLSX is not built and the app does not promise Excel compatibility.
  • The prohibition lists in sections 1 and 13 of TODO.md stay unticked. They are standing prohibitions, to be re-confirmed before release.
  • Whether a date is overdue is decided in the browser, so /api/key-status stays independent of the current date.

Data model (schema version 8; version 9 after section 24)

  • keys: key_number (1–3), name, type (hardware/cloud), created_at. The name is written once and there is no device column.
  • health_checks: checked_at, next_check_due, wallet_software, software_version, software_compatibility_status, software_comments, contact_info_verified, comments, plus the legacy sparrow_version / nunchuk_version. wallet_software and contact_info_verified are free text.
  • key_checks: one row per submission — available, hardware_works, firmware_version, seed_backup_verified, cloud_health_status, cloud_provider_operational, address_path_verified, address_path, comments.
  • tool_checks: removed by section 24. Until then it holds product (free text), bip327_status, slip39_status, with UNIQUE (health_check_id, product).
  • ai_security_checks: provider (free text), status, comments, with UNIQUE (health_check_id, provider).
  • Startup migration rebuilds keys when it still has device, key_checks and ai_security_checks when they use the old layout, and health_checks, key_checks and tool_checks when their CHECK constraints still allow the old values — IKKJE_VURDERT, contact_info_verified IN (0, 1), or one of the fixed product, provider, and wallet-software lists.
  • The table definitions are {table} templates, so a rebuild can create the same table under a temporary name. health_checks is a parent table, so the whole migration runs with PRAGMA foreign_keys = OFF, checks PRAGMA foreign_key_check before committing, and rolls back on any failure.
  • Section 24 adds no table. It drops tool_checks and bumps the version to 9, and the drop needs no rebuild of a parent table.

API

Pages: /, /setup, /health-check, /history. Endpoints: GET /health, GET /api/setup, POST /api/setup, GET /api/keys, GET /api/key-status, GET /api/previous-values, GET /api/health-checks, GET /api/health-checks/<id>, GET /api/alerts, GET /api/export, POST /api/health-checks. Field details are in README.md.

Three things that are easy to break:

  • New static files must be added to STATIC_FILES in app.py, or they return 404.
  • /api/setup, /api/alerts, /api/key-status and /api/previous-values are extra endpoints that are not in the minimum list in section 11. They are noted in section 11 but not in the specification itself.
  • STORAGE_ERRORS in app.py includes RuntimeError, because list_keys() raises it for a partial setup. Keep it there, or a broken database answers with a traceback instead of a 500.

Traps that already cost time — do not repeat

  • with sqlite3.connect(...) does not close the connection. It only commits or rolls back the transaction. On Windows the database file stays locked and TemporaryDirectory.cleanup() fails with WinError 32. Always close explicitly with try/finally; see write_schema() in tests/test_database.py.
  • A required control inside a display:none container still blocks submission. The browser refuses to submit and logs "An invalid form control is not focusable". Hiding a section is not enough: the required flags must be cleared, or the section must render something other than required fields. That is why renderSecurityChecks() shows an empty note instead of cards when the key is unavailable, and why syncUnavailableState() clears softwareCompatibilitySelect.required. Section 24.1 adds Hardware works to that list.
  • Do not assert on the quoting style of the JavaScript. static/health-check.js uses single quotes for constant selectors and template literals where a ${...} substitution is needed. Assert on the selector text, e.g. [name="address_path"].
  • clean_record() in tests/test_alerts.py has a fixed provider list. A test that needs another provider must replace the entry itself.
  • sample_payload() / unavailable_payload() are duplicated in tests/test_validation.py, tests/test_database.py and tests/test_app.py. They build the labels from the key's name and the selected software, so a renamed key works without further changes. Section 24 removes their tool_checks part.
  • tests/test_app.py has three test classes. ServerTestCase starts the server, AppHandlerTests stores the default setup in setUp, RenamedSetupTests stores the renamed one, and KeySetupApiTests leaves the database empty. A new test must go in the class whose setup it needs.
  • Section 19 records one key per submission, so a Check id in the export can cover up to three keys only for rows saved before section 19.

Product clarification

The key's configured name is its product label, and a cloud key's name is also a wallet-software choice. A hardware key name is never a wallet-software choice, and a cloud key records its own name as its software. Section 24 removes the BIP327/SLIP39 rows, so a submission records the security statuses and nothing else.