Repository Issue Activity (beta)

#266 Long rendered responses are duplicated in terminal scrollback because Rich `Live` uses `vertical_overflow="visible"`

#220 `init` command doesn't use `--kb-dir` option

#264 docs(skill): prefer MCP server / CLI over grep for external-agent search

#262 feat(mcp): expose taxonomy browsing + tiered search as an MCP server

#260 feat(cli,agent): expose tiered search + taxonomy via CLI and wire into query/chat agent

#258 feat(search): tiered BM25 search (briefs/summaries/sources) + taxonomy accessors

#256 fix(agent): summary-rewrite falls back to v1 instead of retrying with a smaller prompt on context-window overflow

#235 fix(agent): stream LLM completions to avoid gateway idle-timeout on long compiles

#254 fix(locks): retry atomic_write_bytes on transient PermissionError (WinError 5)

#252 fix(converter): cap doc_name length to avoid Windows path-length errors

#251 ADC Issue: Allow GCP ADC and cloud IAM auth without requiring static VERTEX_AI_API_KEY

#249 fix(indexer): close PageIndex's local SQLite connection so rollback doesn't hit WinError 32

#233 feat(agent): hybrid BM25 full-text search tool for query/chat agent

#247 feat(agent): curb concept/entity sprawl (strict entity types, name-length gate, pending buffer)

#245 feat(agent): opt-in append-note concept_update_mode to avoid full-page rewrites on ingest

#231 feat(cli): add-all command and auto_delete_added_files config

#244 Need a new feature to limit the concept/<concept>.md size and auto spawn <concept>-<child-concept>.md

#203 Bulk ingestion: per-document rewrites of entity/concept pages scale ~O(N²) on large, topically-dense corpora — propose a two-phase (map/reduce) consolidation mode

#242 `get_image` fails with OpenAI-compatible Chat Completions endpoints

#239 feat(cli): configurable insert_mode (fail-fast / fail-at-end) for strict compile-failure handling

#238 openkb 0.4.5 requires litellm==1.87.2 which doesn't exist on PyPI — should pin to 1.98.0 or later.

#237 OpenKB analyzed by Claude, and comparision with other memory systems

#229 fix(compiler): retry transient LLM API timeouts with bounded backoff

#226 concepts-plan prompt embeds an uncapped, uncached listing of every concept and entity page (measured 3.5x per-document input growth over 28 docs)

#225 Compiler writes an unvalidated model non-answer verbatim into wiki/index.md as a document description