v0.8.3
vectorize-io/hindsightv0.8.3Jun 18, 2026by github-actions[bot]
Upgrading?
Release notes summary
A maintenance release focused on bug fixes, performance improvements, and new integrations including Zapier, Continue, Zed, and OpenHands, alongside new features like shared observation scopes and MCP tool annotations.
Key Highlights
- Zapier integration with actions and REST Hook triggers
- Shared observation_scopes keyword for consolidation
- MCP ToolAnnotations for read-only and destructive hints
- Intel XPU local embedding acceleration support
- Zed and OpenHands editor integrations
New Features
- Zapier app integration
- Continue.dev integration via HTTP context provider
- Zed editor integration (MCP context server + recall/retain rule)
- OpenHands integration (native MCP config + recall/retain rule)
- Composio integration (Hindsight memory as custom tools)
- Shared observation scope support
- MCP tool annotations
- Intel XPU detection for local embeddings
- Dry-run fact extraction endpoint (preview mode)
Full Release Notes
## What's Changed * docs: changelog and blog post for v0.8.2 by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2172 * docs: add Memory Defense to the 0.8.2 blog post by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2173 * blog: Cursor persistent memory (editor + CLI in one post) by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2171 * docs(models): register the nous provider so the Models page lists it by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2128 * docs(litellm): replace removed `opinion` fact-type with `observation` by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2198 * fix(search): use effective-time fallback (mentioned_at, occurred_end) for recency scoring by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2197 * Chore (opencode): drop utility re-exports from plugin entry by @mdbenito in https://github.com/vectorize-io/hindsight/pull/2193 * fix(metrics): don't count a client-disconnect cancellation as a failed recall/reflect operation by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2185 * docs(reflect): align reflect_async docstring with read-only tool set by @kubaodias in https://github.com/vectorize-io/hindsight/pull/2200 * feat(consolidation): add "shared" observation_scopes keyword by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2202 * fix(db): widen remaining live bank_id columns to TEXT on PostgreSQL (#2106 follow-up) by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2175 * fix(api): mental model delta refresh (prompt size, JSON, consolidation) by @oldschoola in https://github.com/vectorize-io/hindsight/pull/2170 * feat(api): omit null fields from JSON responses where wire-safe by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2204 * feat(migrations): parallelize tenant schema migrations by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2203 * feat(retain): make structured-unit chunk limit configurable by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2139 * feat(api): dry-run fact extraction endpoint (preview, no persistence) by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2205 * fix(ts-client): accept shared observation scope by @koriyoshi2041 in https://github.com/vectorize-io/hindsight/pull/2207 * blog(gemini-spark): Gemini Spark persistent memory via MCP by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2208 * feat(zapier): add Hindsight Zapier app (actions + REST Hook triggers) by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2119 * feat(continue): add Continue.dev integration via HTTP context provider by @DK09876 in https://github.com/vectorize-io/hindsight/pull/2213 * fix(skill-docs): convert all admonition keywords in the docs→skill generator by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2218 * fix(curation): drop the embedding column from invalidated_memory_units (#2209) by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2210 * feat(api): improve Chinese temporal query parsing by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2220 * perf(api): index memory_links.bank_id on PostgreSQL by @zommiommy in https://github.com/vectorize-io/hindsight/pull/2223 * fix(api): gate dry-run extraction behind the operation precheck by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2211 * fix(docs): deflake memories.py doc example (async-consolidation race) by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2152 * test: fix flaky test-api shard (migration pg0 provisioning race) by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2225 * fix(mcp): give update_memory/invalidate_memory non-empty descriptions by @bluecrayon52 in https://github.com/vectorize-io/hindsight/pull/2215 * test(control-plane): validate t() keys resolve against the i18n catalog by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2232 * feat(tags): officially surface tags_match=exact in UI, docs, and clients by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2230 * fix(migrations): merge two divergent alembic heads by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2234 * test: fix flaky test_llm_trace via leaked global recorder (#2229) by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2231 * revert(worker): trust schemas_with_pending_work() result, drop per-poll re-scan by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2236 * docs: update CLAUDE guidance by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2228 * docs: fix stale documentation links by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2221 * fix(control-plane): expose "shared" observation scope in the Add Document UI by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2212 * blog(obsidian): Chat With Your Obsidian Vault, grounded in your notes by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2237 * feat(composio): add Composio integration (Hindsight memory as custom tools) by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2180 * chore(control-plane): format bank-selector.tsx (lint drift from #2212) by @DK09876 in https://github.com/vectorize-io/hindsight/pull/2250 * Add TypeScript client getVersion helper by @koriyoshi2041 in https://github.com/vectorize-io/hindsight/pull/2252 * Strip reasoning tags on non-structured LLM output (and unclosed blocks) by @grimmjoww in https://github.com/vectorize-io/hindsight/pull/2195 * fix(api): reject empty/whitespace content in dry-run extraction before the LLM call by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2246 * docs(embed): document control center commands by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2151 * docs(paperclip): document dynamicBankId / bankId / user granularity (#1761) by @r266-tech in https://github.com/vectorize-io/hindsight/pull/1803 * Add Python client get_version helper by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2256 * fix(api): defer provider quota reset retries by @oldschoola in https://github.com/vectorize-io/hindsight/pull/2194 * chore(ci): disable Dependabot version updates by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2257 * Honor CODEX_HOME for Codex auth.json location by @mdenil in https://github.com/vectorize-io/hindsight/pull/1874 * fix(openclaw): always session-scope retained documents by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2259 * fix(claude-code-mcp): resolve venv interpreter in the Windows Scripts/ layout (#1758, 3a) by @eldar702 in https://github.com/vectorize-io/hindsight/pull/2066 * claude-code: case-insensitive directoryBankMap matching on Windows by @eskild-boop in https://github.com/vectorize-io/hindsight/pull/2183 * fix: template bank-id path segment in HTTP metric endpoint label (follow-up to #850) by @sireika in https://github.com/vectorize-io/hindsight/pull/2191 * feat(api): add optional MarkItDown OCR support by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2145 * fix(litellm): hard-cap completions with asyncio.wait_for so a hung call can't block forever by @kubaodias in https://github.com/vectorize-io/hindsight/pull/2224 * log(consolidation): show call count + avg for each timing phase by @cdbartholomew in https://github.com/vectorize-io/hindsight/pull/2238 * feat(mcp): add ToolAnnotations (read-only/destructive hints) to MCP tools by @yunanwg in https://github.com/vectorize-io/hindsight/pull/2243 * fix(consolidation): handle single-value source_fact_ids from LLM by @haodonp in https://github.com/vectorize-io/hindsight/pull/2240 * feat(extensions): expose Content-Length on PrecheckContext by @cdbartholomew in https://github.com/vectorize-io/hindsight/pull/2247 * feat(metrics): expose async-operation queue + consolidation backlog as gauges by @MattJackson in https://github.com/vectorize-io/hindsight/pull/1987 * fix(api): apply bank-config disposition + mission overlay in list_banks by @PathKnower in https://github.com/vectorize-io/hindsight/pull/2101 * fix(openclaw): strip runtime metadata from memory content by @de1tydev in https://github.com/vectorize-io/hindsight/pull/1968 * fix(mcp): omit reflect tool_trace/llm_trace from responses by default by @yunanwg in https://github.com/vectorize-io/hindsight/pull/2242 * Add Gemini service tier config by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2251 * feat(embeddings): detect Intel XPU for local embedding acceleration by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2260 * docs: document MCP trace and precheck content length by @r266-tech in https://github.com/vectorize-io/hindsight/pull/2264 * test(openclaw): fix agent_end hook tests after #1968 context change by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2266 * test(retain): fix broken quota-defer test config mock by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2265 * fix(retain): stop dropping chunks when an oversized doc splits into multi-chunk sub-batches by @nicoloboschi in https://github.com/vectorize-io/hindsight/pull/2269 * Hide Windows netstat port probes by @koriyoshi2041 in https://github.com/vectorize-io/hindsight/pull/2263 * Update integrations.json by @yugandhar-maram in https://github.com/vectorize-io/hindsight/pull/2268 * docs(observations): stop claiming a per-observation freshness trend by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2271 * feat(zed): add Zed editor integration (MCP context server + recall/retain rule) by @DK09876 in https://github.com/vectorize-io/hindsight/pull/2153 * blog(freshness): Freshness-Aware Memory — knowing when a belief has gone stale by @benfrank241 in https://github.com/vectorize-io/hindsight/pull/2267 * feat(openhands): add OpenHands integration (native MCP config + recall/retain rule) by @DK09876 in https://github.com/vectorize-io/hindsight/pull/2276 * chore(deps): fix critical/high Dependabot alerts by @dcbouius in https://github.com/vectorize-io/hindsight/pull/2278 * fix(tests) backsweep migration test pg0 isolation by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2282 * fix(tests) S3 testcontainers port mapping race by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2283 * fix(mcp): restore custom tool instructions by @Sanderhoff-alt in https://github.com/vectorize-io/hindsight/pull/2288 * feat(metrics): instrument async worker completion path with operation metrics by @kubaodias in https://github.com/vectorize-io/hindsight/pull/2253 ## New Contributors * @mdbenito made their first contribution in https://github.com/vectorize-io/hindsight/pull/2193 * @kubaodias made their first contribution in https://github.com/vectorize-io/hindsight/pull/2200 * @koriyoshi2041 made their first contribution in https://github.com/vectorize-io/hindsight/pull/2207 * @zommiommy made their first contribution in https://github.com/vectorize-io/hindsight/pull/2223 * @bluecrayon52 made their first contribution in https://github.com/vectorize-io/hindsight/pull/2215 * @mdenil made their first contribution in https://github.com/vectorize-io/hindsight/pull/1874 * @eldar702 made their first contribution in https://github.com/vectorize-io/hindsight/pull/2066 * @eskild-boop made their first contribution in https://github.com/vectorize-io/hindsight/pull/2183 * @yunanwg made their first contribution in https://github.com/vectorize-io/hindsight/pull/2243 * @haodonp made their first contribution in https://github.com/vectorize-io/hindsight/pull/2240 * @MattJackson made their first contribution in https://github.com/vectorize-io/hindsight/pull/1987 * @PathKnower made their first contribution in https://github.com/vectorize-io/hindsight/pull/2101 * @yugandhar-maram made their first contribution in https://github.com/vectorize-io/hindsight/pull/2268 **Full Changelog**: https://github.com/vectorize-io/hindsight/compare/v0.8.2...v0.8.3
Related Repositories
public-apis/public-apis★ 491,660
A collective list of free APIs
NousResearch/hermes-agent★ 256,513
The agent that grows with you
yt-dlp/yt-dlp★ 200,929
A feature-rich command-line audio/video downloader
microsoft/markitdown★ 191,817
Python tool for converting files and office documents to Markdown.
Significant-Gravitas/AutoGPT★ 187,717
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.