Keysat
2e70b34592
Phase 1 Workstream D. Lets the Architect ground the thesis in REAL recurring LP objections without any LP identity reaching the Claude API. Layered, defense-in-depth, fail-closed by construction (docs/redaction-rehydration.md). backend/redaction/: - scrub.py: the leak-proof core. Drops Tier-1 (labelled/structured account/wire/SSN/ IBAN/SWIFT/passport, separator-tolerant); tokenizes known LP entities (dictionary from the canonical layer, unicode-folded + hyphen-extended) and structured PII (emails, scheme-less/social URLs, intl+ext phones, currency-cued amounts, ISO/worded/numeric/ quarter dates, addresses, bare long digit runs); pre-neutralizes injected [TYPE_N] strings; single-pass rehydrate; metadata-only audit logging (the pseudonym map is the de-anon key — local-only, never logged/sent). Hardened across THREE adversarial leak-hunts (worded/coded amounts, intl phones, NFD/ligature/zero-width names, slash/ comma SSN, SWIFT, alpha-prefixed accounts, substance-preserving false-positive fixes). - client.py: Boundary — one scrub/rehydrate contract, SCRUB_BACKEND=local (default) or gateway (Spark Control /scrub + /rehydrate). Fails closed (db_path required; dictionary build errors propagate; strict rehydrate returns tokenized-not-de-anon text). - test_scrub_leak.py, test_reidentification.py: golden-file leak + re-identification suites (synthetic only, guardrail #9), regression-locking every leak-hunt vector. backend/mcp/architect_grounding.py: the flow — retrieve (local) -> minimize-first (local Qwen) -> scrub (+ local-Qwen NER backstop for unknown names) -> Claude over the de-identified register only -> re-hydrate locally -> human review. FAILS CLOSED if the local model is unreachable or a hallucinated token appears. test_grounding_boundary.py proves nothing sensitive reaches Claude and the three fail-closed paths. server.py: POST /api/architect/ground (admin) wires retrieval -> ground_objections. docker_entrypoint.sh: SCRUB_BACKEND (default local). docs/spark-control-scrub-endpoints.md: the gateway handover spec (Option 1 — caller supplies the entity dictionary). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
23 lines
1.3 KiB
TypeScript
23 lines
1.3 KiB
TypeScript
import { VersionInfo } from '@start9labs/start-sdk'
|
|
|
|
// The Architect grounding boundary (Phase 1 Workstream D): the redaction / re-hydration
|
|
// privacy gate so the Architect can ground its thesis in REAL LP feedback without sending
|
|
// any LP identity to Claude. A local-Qwen summary minimizes first, a deterministic scrubber
|
|
// (hardened across three adversarial leak-hunts) tokenizes residual identifiers, a local-Qwen
|
|
// NER pass backstops unknown names, Claude reasons only over the de-identified register, and
|
|
// the draft is re-hydrated locally for human review. Fails CLOSED if the local model is
|
|
// unreachable. SCRUB_BACKEND=local (default) or gateway (Spark Control /scrub + /rehydrate,
|
|
// once built). No data migration.
|
|
export const v_0_1_0_55 = VersionInfo.of({
|
|
version: '0.1.0:55',
|
|
releaseNotes: {
|
|
en_US: [
|
|
'Adds the Architect grounding boundary: the Architect can now pressure-test the thesis',
|
|
'against real recurring LP objections WITHOUT any LP name, email, amount, or other',
|
|
'identifier reaching Claude — everything is de-identified locally first and the draft is',
|
|
'restored on the server for your review. Sovereignty by construction; fails closed.',
|
|
].join(' '),
|
|
},
|
|
migrations: { up: async () => {}, down: async () => {} },
|
|
})
|