Status: pre-launch. This page describes target design; this document does not establish a deployed service.

DEV-05 -- Add A Cortex Hook

Source: a private bridge module that connects the Rosetta chain to organism memory. The source path is not reproduced here.

What The Bridge Does Today

The bridge module connects the Rosetta chain to organism memory:

• Writes a markdown trace to a private memory store.
• Retrieves prior traces by similarity.
• Semantic recall uses sentence-transformers if installed.
• Keyword recall is the fallback when semantic tooling is unavailable.

Hook Points

Hook	Purpose	Guard
before L2 exploration	recall similar traces for context	never treat recall as source truth
after chain completion	write trace receipt	include chain path, success, action, full JSON trace
before apu.bot's Cortex-gardener cadence	expose accumulated traces	apu.bot suggests; it does not silently rewrite
before authority review	attach precedent/context	A Natural Person signs their own typed private act; public-DAV consequence requires a complete valid bound PRISM decision receipt from the signer(s) required by applicable law, the entity's governing instrument, or a valid competent mandate binding the exact consequence. PRISM records and verifies only; the hook remains unsigned context

Add A New Hook

1. Decide whether the hook reads memory, writes memory, or both.
2. Use Path-resolved repo paths; avoid hard-coded absolute user paths.
3. Preserve append-only behavior for receipts and traces.
4. If the hook changes lineage or archives material, apu.bot must stage an unsigned proposal rather than write directly. Apply it only after the relevant typed authority receipt is valid.
5. Add a test around empty memory, missing optional dependency, and malformed trace input.

Non-goals

• Cortex is not the signer.
• Cortex is not the public claim authority by itself.
• Cortex recall is context, not proof.