supermemory · Issues· 106 open

[Bug]: v0.0.8 self-hosted memory generation fails with No model provider configured while doctor detects OpenAI

Claude memory tool: path → customId normalization collides, silently overwriting unrelated memory files

`forgetMemoryRequest` drops its 30s timeout whenever a caller passes a signal

Local console (#memory) broken on non-localhost hosts: no auth sent + wrong graph endpoint

[Bug]: SUPERMEMORY_EMBEDDING_PROVIDER and related env vars are ignored — embedding hardcoded to local

self-hosted v0.0.6 Darwin arm64 binds all interfaces; no documented loopback override

[Feature Request] Self-hosted: full export/list API for extracted memories (currently only search-based recall)

Self-hosted: mid-file edit to an upserted document re-embeds nearly all chunks (position-sensitive chunking) — ~21 min CPU per edit on 2 vCPU with local embeddings

Where is the source for supermemory-server?

Supermemory v0.0.8 |Cannot load documents and stats that's NOT from localhost

Proposal: Add SDK idempotency keys for safe memory write retries

RFC: Introduce a Working Memory layer in the AI SDK for low-latency local context retrieval

[Feature Request] Memory governance layer — PII redaction, context poisoning defense, and audit trail for memory retrieval`

MCP: a transient `/v3/session` failure is reported to the client as an invalid token

Self-hosted server: cron failsafe logs neither document IDs nor timestamps