Repository Issue Activity (beta)

#1206 Vercel AI SDK middleware silently drops tool-call and tool-result messages when saving conversations

#1205 Dead code in search threshold schemas: unreachable `v === undefined` checks and no-op `.transform(Number)`

#1203 Self-hosted v0.0.3: documents over ~64 KiB permanently wedge the ingest queue (128 KiB storage-value limit) — never done/failed, undeletable (409 loop), no cancel

#1200 memory-graph should expose document/source labels and hover popover layering

#1191 fix(ai-sdk): use z.coerce.number() for limit/offset params to handle string inputs from LLMs

#1027 BUG: AttributeError when profile is None in _retrieve_memories()

#1184 Plugin transcript parser drops every line after the first in multi-line messages

#1030 [Feature] Add 'listMemories' tool to MCP server for enumerating user memories

#1175 YouTube URL detection matches domains by substring

#1180 Harden claude-auto-fix-ci.yml workflow permissions

#1177 Self-hosted v0.0.3: encrypted snapshot OOMs (RangeError in node:crypto) past ~150 MB DB, orphaning docs into a permanent retry-cron loop (plus lossy deletion and a runaway memory agent)

#1173 Social URL detection matches domains by substring

#1169 Feedback request: AI PR descriptions for fast-moving AI teams

#1102 Publish a Windows-native server binary (supermemory-server-windows-x64) + add a win32 path to the installer/launcher

#1104 Self-hosted: make the local embedding model configurable (hard-coded bge-base-en-v1.5 breaks non-English recall)

#1152 Self-hosted XLSX upload indexes ZIP/OpenXML bytes instead of spreadsheet cell text

#1151 Self-hosted file uploads fail extraction: /files prefix is passed as local storage key

#985 Expose privacy-safe memory retrieval receipts

#1148 Support `HOST` env var to bind server to loopback only

#1140 Explicit `invalid_condition` support + self-host offline + pricing transparency

#1124 BUG: OAuth redirection loop / repeated organization selection during MCP client authorization

#1093 Self-hosted server: unbounded memory growth in local embedding workers under continuous ingestion (reaches 22–50 GB)

#1096 self-hosted memory agent fails with Ollama due to Responses item_reference input

#1092 Memory agent fails with OpenAI-compatible endpoints that return reasoning_content (e.g. Xiaomi MiMo)

#1103 Self-hosted v0.0.2: search always returns total:0 on a store upgraded from v0.0.1 (fresh store works)