A 429 usage_limit_reached response during verify_connection() caused the
server to refuse to start entirely. Quota exhaustion is not a configuration
error — the server should start and serve retain/recall requests normally,
it just can't make LLM calls until the quota resets.
Co-authored-by: Marco Rutsch <marco@rutimka.de>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Add optional filter_mcp_tools() method to OperationValidatorExtension.
Called during tools/list after bank-level mcp_enabled_tools filtering.
Extensions can override to hide MCP tools per-user-per-bank based on
access policies. Default returns all tools unchanged.
- Add filter_mcp_tools to OperationValidatorExtension with default pass-through
- Wire into _get_enabled_tools in _apply_bank_tool_filtering
- Move _ALL_TOOLS to mcp_tools.py to avoid circular import (re-exported from mcp.py)
- Fail-open: if filter raises, log warning and return unfiltered tools
- Enforce ceiling: validator can narrow but never expand beyond bank config
- Add 8 tests: default, filtering, empty set, integration, composition,
can't-add-tools, exception fail-open, no-validator passthrough
* fix: parse query params from base_url in OpenAI embeddings client
The OpenAI-compatible LLM provider already parses query parameters
(e.g. ?api-version=xxx for Azure OpenAI) from the base_url and passes
them as default_query to the OpenAI client. However, the OpenAI
embeddings provider did not do this, causing Azure OpenAI embeddings
to fail with 404 errors at runtime.
This applies the same URL parsing logic from the LLM provider to the
embeddings provider, enabling Azure OpenAI embeddings to work correctly.
* ci: add workflow to build fork Docker image
* ci: add slim image build (no local models)
* ci: remove fork build workflow per review request
---------
Co-authored-by: Antoine Khater <ak@ptgroup.eu>
* fix(claude-code): implement tool_choice support for forced tool calls
The call_with_tools() method now properly handles the tool_choice parameter
to force specific tool calls. Previously, the parameter was accepted but ignored,
causing the reflect agent to fail when trying to force specific tools on each
iteration.
Fixes#732
Changes:
- When tool_choice forces a specific function: filter allowed_tools to only
that tool (with mcp prefix) and add a strong system prompt instruction
- When tool_choice is 'required': add instruction that model must call at
least one tool
- When tool_choice is 'none': clear allowed_tools and mcp_servers to disable
all tools
- When tool_choice is 'auto' (default): no change (existing behavior)
This matches the approach used in the OpenAI provider while adapting to the
Claude Agent SDK's lack of native tool_choice parameter by using allowed_tools
filtering and system prompt instructions.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* style: fix ruff formatting in alembic migration
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat: add max_observations_per_scope bank config
Adds a configurable limit on the number of observations per tag scope.
When the limit is reached, consolidation only updates/deletes existing
observations — no new ones are created. Enforcement is done via a
constrained Pydantic response model (max_length on creates list) so the
LLM structurally cannot exceed the limit, plus prompt guidance.
- Config: HINDSIGHT_API_MAX_OBSERVATIONS_PER_SCOPE (-1 = unlimited)
- Reorder action execution: deletes → updates → creates
- Dynamic _ConsolidationBatchResponse with max_length constraint
- Prompt CAPACITY CONSTRAINT section when near/at limit
- Observations with no tags skip the limit entirely
- Control plane UI field + docs
* fix: strengthen max_observations tests with mock LLM + defensive truncation
- Rewrite integration tests to use MockLLM with deterministic responses
(one observation per fact) instead of relying on real LLM behavior
- Add defensive truncation in _consolidate_batch_with_llm as belt-and-
suspenders — catches LLM providers that ignore JSON schema max_length
- Tests now assert exact counts, not just upper bounds
The stats endpoint JOINs memory_links to memory_units just to filter
by bank_id. With 8.2M+ links per bank this takes 18+ seconds, and
the control plane polls every 10s — perpetually blocking the server.
Add bank_id column directly to memory_links so the query can filter
on ml.bank_id instead of mu.bank_id, letting Postgres push the filter
down before the JOIN.
- Migration: add bank_id TEXT NOT NULL, backfill from memory_units
- All 4 INSERT paths (temporal, semantic, entity, causal) now write bank_id
- Stats query filters on ml.bank_id instead of mu.bank_id
* fix(migrations): use HINDSIGHT_API_MIGRATION_DATABASE_URL when set
Session-level advisory locks are broken when the database URL goes
through PgBouncer in transaction mode: the backend connection is
returned to the pool on COMMIT, orphaning the lock, so multiple pods
can simultaneously run migrations for the same schema.
When HINDSIGHT_API_MIGRATION_DATABASE_URL is set, use it for both
the advisory lock connection and the Alembic run. Callers should
point this at the direct PostgreSQL endpoint (bypassing the pooler)
so the session-level lock is held for the full migration duration.
* refactor(migrations): move MIGRATION_DATABASE_URL to standard config
Wire HINDSIGHT_API_MIGRATION_DATABASE_URL through HindsightConfig
instead of reading os.getenv() directly in migrations.py. Add the
field to the dataclass, from_env(), log_config(), all call sites,
.env.example, and the configuration docs page.
* fix: update test mocks for migration_database_url kwarg and regenerate docs skill
---------
Co-authored-by: Nicolò Boschi <boschi1997@gmail.com>
* fix: silence noisy google_genai.models INFO logging
The google-genai SDK logs "AFC is enabled with max remote calls: 10"
at INFO level on every initialization. Set its logger to WARNING.
* fix: regenerate docs skill in release-integration script
The release script generates changelog/SDK pages but never re-ran
generate-docs-skill.sh, causing CI to fail with out-of-sync skill
files after every integration release. Now it regenerates the skill
and includes the output in the release commit.
Also adds the missing ag2 skill files from the latest release.
* fix(migration): use IF EXISTS when dropping chunk FK constraint
The migration unconditionally dropped memory_units_chunk_fkey, but
depending on the order in which migrations were applied the constraint
may not exist. Use raw SQL with IF EXISTS so the drop is safe regardless.
* fix(migration): make chunk FK add idempotent with DO block
The previous fix only handled the DROP side with IF EXISTS. The ADD side
could still fail with DuplicateObject when the FK already existed on a
schema that was provisioned after the base migration ran.
Wrap the ADD CONSTRAINT in a DO block to catch duplicate_object and
continue, making the migration fully idempotent in both directions.
Port fixes from #461 (claude_code_llm) to codex_llm:
- Replace json.dumps(result) with result.model_dump_json() for Pydantic models to fix TypeError during consolidation
- Wrap record_llm_call tracing block in try/except so logging failures never propagate to retry handler
Co-authored-by: Marco Rutsch <marco@rutimka.de>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat: add audit log for feature usage tracking
Add full auditability for all mutating and core API operations across
HTTP, MCP, and system (worker) transports. Audit entries record raw
request/response as JSONB, timing (started_at/ended_at), action, and
transport type.
Backend:
- New audit_log table with JSONB columns for expandability without
future migrations (merge migration of 3 existing heads)
- AuditLogger with fire-and-forget writes via asyncio.create_task
- @audited decorator on 28 HTTP route handlers
- MCP tool audit wrapping for 16 auditable tools
- Worker task execution wrapped with audit_context
- List endpoint with action, transport, date range filters + pagination
- Stats endpoint with per-day counts for charting
- Configurable retention sweep (concurrent-safe DELETE)
Config (env-only, static):
- HINDSIGHT_API_AUDIT_LOG_ENABLED (default: false)
- HINDSIGHT_API_AUDIT_LOG_ACTIONS (comma-separated allowlist, empty=all)
- HINDSIGHT_API_AUDIT_LOG_RETENTION_DAYS (default: -1, keep forever)
Control Plane:
- New "Audit Logs" tab on bank configuration page
- Line chart showing request volume (today/7d/30d) with action filter
- Filterable table with action, transport, date range filters
- Paginated list with detail dialog showing raw request/response JSON
Tests:
- 13 tests covering list, filters, pagination, stats, disabled mode,
action allowlist, and ordering
* fix: split 3-way merge migration into two 2-way merges
Alembic doesn't support 3-parent merge migrations. Split into a no-op
merge of 2 heads (b1c2d3e4f5g6) followed by the audit_log table
migration merging the third head.
* fix: correct merge migration to merge actual 2 heads
The original analysis incorrectly identified 3 heads. There were only 2
(a3b4c5d6e7f8 and c8e5f2a3b4d1). Remove the unnecessary intermediate
merge migration and fix the audit_log migration to merge these 2 heads.
* fix: use 'heads' instead of 'head' in migration runner
Alembic's upgrade('head') fails when multiple heads exist (e.g. from
namespace package overlaps between hindsight-api and hindsight-api-slim).
Using 'heads' (plural) handles this gracefully by upgrading all branches.
* chore: regenerate OpenAPI spec with audit log endpoints
* chore: regenerate TypeScript client and docs skill OpenAPI spec
Python and Go clients still need regeneration (requires Docker).
* chore: regenerate all client SDKs (Python, Go, TypeScript)
Adds generated audit log API clients for Python (audit_api.py),
Go (api_audit.go), and TypeScript client type updates.
- Add 'ark' and 'volcano' as valid LLM providers (both are aliases for Volcano Engine)
- Set default model to 'doubao-pro-32k' for both providers
- Add them to OpenAICompatibleLLM provider list
- Exclude from json_object response format support
Co-authored-by: yishun.eason <yishun.eason@bytedance.com>
* docs(python-client): improve pydoc strings for async-first usage and low-level API access
- Class docstring now clearly documents async-first pattern: a* methods
preferred, sync wrappers for scripts/REPLs only
- Every sync method docstring points to its async counterpart
- Every async method docstring says "preferred"
- Expose 10 low-level API properties (documents, entities, operations,
webhooks, monitoring, etc.) so agents/users can discover the full API
surface without guessing at _-prefixed internals
- Add missing API parameters: tag_groups (recall/reflect), fact_types,
exclude_mental_models, exclude_mental_model_ids (reflect),
observation_scopes/strategy (retain items), background (create_bank)
- Fix areflect missing include_facts param that sync reflect already had
- Sync recall/reflect now delegate to async counterparts (no logic duplication)
* style(retain): format long function call arguments one-per-line
* feat(retain): delta retain — skip LLM re-extraction for unchanged chunks on upsert
When upserting a document (same document_id), instead of deleting all
facts and re-extracting from scratch, compare chunk content hashes
and only process changed/new chunks. Unchanged chunks keep their
existing facts, entities, and links.
- Add content_hash column to chunks table (migration b3c4d5e6f7a8)
- Add chunk delta comparison functions in chunk_storage.py
- Add delta_mode to fact_storage.handle_document_tracking (skip full delete)
- Add update_memory_units_tags for propagating tag changes to existing facts
- Refactor orchestrator into _try_delta_retain and _full_retain paths
- Automatic fallback to full retain for pre-migration data or all-changed scenarios
- Fix ty type error in metrics.py (resource module import on Windows)
- 16 new tests covering entities, links, tags, metadata, edge cases
* refactor(retain): deduplicate delta and full retain paths
Extract shared _insert_facts_and_links() and _extract_and_embed()
functions used by both the full retain and delta retain paths.
Remove delta_mode flag from handle_document_tracking — delta path
uses dedicated upsert_document_metadata() instead.
* chore: regenerate clients, openapi spec, and lockfile
* chore: regenerate docs skill
* docs: add supported platforms section and Windows installation guide
Adds a platform compatibility table (Linux, macOS, Windows) and a
dedicated Windows setup section with step-by-step instructions for
installing PostgreSQL + pgvector and running Hindsight natively.
Follows up on #699 which added Windows native support.
Also fixes a ty type-check error in metrics.py for the conditional
resource module import.
* chore: sync generated clients and lock file after #699
Regenerate client SDKs to pick up ValidationError model changes
and update uv.lock with platform-specific uvloop/winloop deps.
* docs: update Windows section — pg0 now supports Windows
pg0 v0.12.0 added Windows support, so embedded DB works everywhere.
Restructure Windows section to show simple install-and-run first,
with external PostgreSQL as an optional alternative.
* chore: sync generated docs skill and openapi references
FastAPI generates the ValidationError schema with only loc, msg, and
type, but Pydantic v2 actually returns input, ctx, and url as well.
Generated clients with strict JSON decoding (Go's DisallowUnknownFields)
cannot parse real 422 responses — the actual validation message gets
replaced by a confusing JSON decoding error.
- Patch the OpenAPI schema in create_app() to add input, ctx, url
- Regenerate spec and Go client
* feat: Windows native support — run Hindsight without Docker on Windows
Four compatibility fixes that allow Hindsight to run natively on Windows
with an external PostgreSQL + pgvector installation:
1. **pyproject.toml**: Conditional event loop dependency
- `winloop` on Windows (sys_platform == 'win32')
- `uvloop` on Linux/macOS (sys_platform != 'win32')
2. **main.py**: winloop integration via `winloop.install()`
- Patches asyncio event loop policy globally before uvicorn starts
- uvicorn sees "asyncio" but runs winloop underneath (same perf as uvloop)
- Falls back to default asyncio if winloop unavailable
3. **metrics.py**: Guard `resource` module import
- `resource` is Unix-only (getrusage, getrlimit)
- Conditional import with None fallback
- Skip process metrics collection on Windows
4. **fact_storage.py**: Cross-platform strftime
- `%-d` (no-padding day) is glibc-only, fails on Windows
- Replaced with `%d` + `.replace(" 0", " ")` for same output
## Windows Setup Guide
### Prerequisites
- Python 3.11+
- PostgreSQL 17 with pgvector extension
- Ollama (for local embeddings) or external embedding provider
### Install PostgreSQL + pgvector on Windows
```bash
winget install PostgreSQL.PostgreSQL.17
# Build pgvector from source (requires Visual Studio Build Tools)
git clone https://github.com/pgvector/pgvector.git
# In x64 Native Tools Command Prompt:
set PGROOT=C:\Program Files\PostgreSQL\17
nmake /F Makefile.win
nmake /F Makefile.win install
# Enable extension
psql -U postgres -d hindsight -c "CREATE EXTENSION IF NOT EXISTS vector;"
```
### Install and Run Hindsight
```bash
pip install -e ".[embedded-db]"
# Set environment variables
set HINDSIGHT_API_LLM_PROVIDER=openai
set HINDSIGHT_API_LLM_API_KEY=your-api-key
set HINDSIGHT_API_LLM_BASE_URL=https://your-llm-endpoint/v1
set HINDSIGHT_API_LLM_MODEL=your-model
set HINDSIGHT_API_DATABASE_URL=postgresql://postgres@localhost:5432/hindsight
set HINDSIGHT_API_EMBEDDING_PROVIDER=ollama
set HINDSIGHT_API_PORT=8889
hindsight-api
```
Data persists in PostgreSQL on your local disk — survives reboots,
updates, and anything that would wipe a Docker volume.
Tested on Windows 11 with PostgreSQL 17.9, pgvector 0.8.2,
Python 3.11, RTX 5080 (CUDA embeddings + reranking).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: handle strftime ValueError on Windows in fact_storage
The strftime call on occurred_start/occurred_end can raise ValueError
on Windows when the datetime object has unexpected format properties.
Wrap in try/except to gracefully skip date signal rather than crash
the entire retain batch.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: control plane UI fixes for recall and data view
- Sanitize NaN cross-encoder scores to 0.0 in reranking pipeline
(Pydantic serializes NaN as JSON null, breaking UI score display)
- Add null-coalesce for score in search debug view to prevent crash
- Switch data view text filter from debounced onChange to Enter key
(avoids slow ILIKE queries on every keystroke for large banks)
- Show loading spinner in search icon during filter requests
- Preserve search/tag filters when clicking "Load more"
* chore: sync generated files after rebase
Adds a proper 'none' provider option so users can run Hindsight as a
chunk store with semantic search but without any LLM dependency, replacing
the hacky workaround of setting provider to 'mock'.
When HINDSIGHT_API_LLM_PROVIDER=none:
- Retain automatically uses chunks mode (no fact extraction)
- Recall works normally (semantic search, BM25, graph retrieval)
- Reflect returns HTTP 400 with clear error message
- Consolidation/observations are disabled
- Mental model refresh returns HTTP 400
- No API key required
* feat(reflect): make source facts in search_observations configurable
The recent fix (#669) hardcoded include_source_facts=False in
search_observations to prevent context overflow. This makes it
configurable via HINDSIGHT_API_REFLECT_SOURCE_FACTS_MAX_TOKENS
(env/tenant/bank), defaulting to -1 (disabled).
- -1: source facts disabled (current behavior, default)
- 0: source facts enabled with no token limit
- >0: source facts enabled with a token budget
* docs: add reflect_source_facts_max_tokens to configuration reference
* fix: update configurable fields count in tests and regenerate docs skill
* feat: add LiteLLM LLM provider for Bedrock and 100+ providers
Add a new `litellm` LLM provider that uses the LiteLLM SDK for chat
completions and tool calling, enabling AWS Bedrock and 100+ other
providers for Hindsight's core engine (retain, recall, reflect).
- New LiteLLMLLM provider in engine/providers/litellm_llm.py
- Registered in factory, valid providers list, and no-api-key set
- Refactored API key validation to use requires_api_key() helper
- Added boto3 dependency for Bedrock auth
- Updated docs: configuration, models, monitoring, providers grid
* feat: add bedrock as first-class LLM provider alias
Add `bedrock` as a dedicated provider name that auto-prepends the
`bedrock/` prefix to model names and delegates to LiteLLMLLM under
the hood. This makes Bedrock support more discoverable — users set
`HINDSIGHT_API_LLM_PROVIDER=bedrock` with plain Bedrock model IDs.
* test: add Bedrock to CI provider tests
- Add bedrock/us.amazon.nova-lite-v1:0 to MODEL_MATRIX in test_llm_provider.py
- Add AWS credential check in should_skip_provider()
- Pass AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION_NAME secrets to test-api job
- Update default bedrock model to amazon.nova-2-lite-v1:0
* fix: regenerate docs skill files and bump memory test timeout
- Regenerate skills/hindsight-docs references after docs changes
- Bump test_llm_provider_memory_operations timeout to 600s for slower
providers like Bedrock via LiteLLM
* test: skip bedrock lite models in memory operations test
Nova Lite has a 10K output token limit which is too low for fact
extraction (requires 64K). The api_methods test (completion, tools,
structured output) already validates the provider works correctly.
* test: use Nova Pro for bedrock CI tests to cover full memory pipeline
Nova Lite only supports 10K output tokens, too low for fact extraction.
Switch to Nova Pro which supports the full 64K output needed for
retain/reflect operations. This ensures bedrock is tested on all
Hindsight functionalities, not just basic API methods.
* test: switch bedrock CI to Nova 2 Lite (supports 64K output tokens)
Nova v1 models (Pro, Lite) have a 10K output token limit which is
too low for fact extraction. Nova 2 Lite supports 64K+ output tokens,
enabling full memory pipeline testing (retain + reflect).
MCP tool bridges sometimes serialize JSON arrays as strings during
transport, e.g. '["a", "b"]' arrives as the literal string '["a", "b"]'
instead of a native JSON array. This causes Pydantic to reject the
input with a validation error.
Add defensive coercion at two layers:
1. HTTP API (http.py): Pydantic field_validator on MemoryItem.tags
with mode="before" that parses JSON strings back into lists.
2. MCP tools (mcp_tools.py): Same coercion in build_content_dict
before tags reach the Pydantic model.
A plain non-JSON string is wrapped in a single-element list.
Correctly-formatted input is passed through unchanged.
Co-authored-by: Philipp <philipp@Philipps-MacBook-Pro.local>
Expose the named retain strategy on the MCP retain tool, matching the
HTTP API's per-item strategy support. This allows MCP clients (Claude
Code, Claude Desktop, etc.) to specify extraction behavior per memory:
strategy: "exact" → verbatim storage, no LLM processing
strategy: "verbose" → detailed extraction
strategy: "concise" → default compressed extraction
Strategies are defined in bank config under retain_strategies.
Unknown strategy names are logged and ignored (bank default applies).
Changes:
- Add strategy param to both retain function signatures (with/without bank_id)
- Add strategy to build_content_dict
- Strategy is set in the content dict, which the engine already handles per-item
Co-authored-by: Philipp <philipp@Philipps-MacBook-Pro.local>
* fix: return metadata in recall responses (#674)
Metadata stored during retain was never retrieved during recall.
Add metadata to all SQL SELECT queries, the RetrievalResult dataclass,
ScoredResult.to_dict(), and MemoryFact construction in the recall pipeline.
* test: add metadata round-trip test for retain→recall
Replace placeholder metadata test with one that actually passes
metadata via retain_batch_async and asserts it is returned on recall.
* fix: parse metadata JSON string from database in MemoryFact
asyncpg may return JSONB columns as strings. Add a field_validator
to MemoryFact.metadata to handle JSON string deserialization.
* security: exclude litellm 1.82.8 (supply chain compromise)
litellm 1.82.8 on PyPI contains a malicious .pth file that
automatically steals credentials on Python startup (no import needed).
See: https://github.com/BerriAI/litellm/issues/24512
Our Docker images ship 1.82.6 and are unaffected, but the open version
constraints (>=1.0.0, >=1.40.0) would allow resolving to 1.82.8 on
fresh installs or lockfile refreshes.
* security: cap litellm at <=1.82.6 (1.82.7 also compromised)
* chore: regenerate uv.lock and openapi spec
* fix: update test to match claude-haiku-4-5 default model name and regenerate docs skill
* chore: fix ruff formatting in generate_changelog.py
* feat(openclaw): remove hardcoded default models, rely on Hindsight API defaults
* feat(claude-code): remove hardcoded default models, rely on Hindsight API defaults
* feat(claude-code,docs): remove hardcoded default models from claude-code integration and docs
* feat: use claude-haiku-4-5 as default Anthropic model
search_observations in the reflect agent hardcoded include_source_facts=True
with max_source_facts_tokens=-1 (unlimited). For banks with many observations
backed by thousands of facts, a single tool call could produce 300K+ tokens,
exceeding the default 100K context budget and causing forced synthesis with
an empty 'Retrieved Data' section.
The reflect agent synthesizes from observations, not raw backing facts.
Disable source facts to keep payloads proportional to observation count
(~6K vs ~310K in the reporter's case).
The consolidation path already has configurable source fact limits (PR #509,
v0.4.17). The reflect path was not updated.
Fixes#668
Co-authored-by: Kagura Chen <daniyuu19@sjtu.edu.cn>
* fix(entity_resolver): prevent _pending_stats/_pending_cooccurrences memory leak
Add discard_pending_stats() to EntityResolver to clean up both pending dicts
for the current task key. Call it at the start of each _run_db_work attempt so
that exceptions between accumulation and flush_pending_stats() — including
deadlock retries — never leave stale entries keyed by recycled task IDs.
Fixes#660
* test(entity_resolver): add unit tests for discard_pending_stats()
Covers: clears both dicts for current task, is idempotent when empty,
and does not touch entries belonging to other task keys.
No database required — purely in-memory logic.
* test: add unit tests for pg_trgm auto-detection and ValidationResult.accept_with() enrichment
Two recent PRs landed without dedicated tests:
- #626/#649 (pg_trgm fallback in EntityResolver): add 5 mocked unit tests
covering the trigram→full fallback, single-check guarantee, and sticky
downgrade behaviour.
- #639 (accept_with() enrichment): add 7 pure unit tests for the factory
method plus 5 integration tests verifying the engine applies enriched
contents (retain) and tags/tag_groups (recall) returned by validators.
Also verifies RecallContext carries tag filter state.
* fix: remove 504 from reflect OpenAPI spec to fix progenitor Rust client build
progenitor-impl-0.11.2 panics with `assertion failed: response_types.len() <= 1`
when an endpoint declares more than one response type. PR #643 added
`responses={504: ...}` to the reflect decorator, which injected a second
response type into the generated OpenAPI spec and broke the Rust client build.
Remove the `responses=` kwarg — the 504 is still raised at runtime via
JSONResponse(status_code=504), it just won't appear in the OpenAPI schema.
Regenerate openapi.json accordingly.
* chore: sync generated files and ruff formatting (lint + docs skill)
On managed PostgreSQL services (e.g. Azure Flexible Server), the pg_trgm
extension may not be available, causing two failures:
1. Migration c1a2b3d4e5f6 crashes on CREATE EXTENSION
2. Even if migration is bypassed, the default 'trigram' entity lookup
strategy uses the % operator which requires pg_trgm, causing retain
background tasks to fail silently
Changes:
- Migration now gracefully skips pg_trgm and index creation if the
extension cannot be loaded
- EntityResolver auto-detects pg_trgm availability on first use and
falls back to 'full' lookup strategy with a warning log
Co-authored-by: coder999999999 <coder999999999@users.noreply.github.com>
Validators can now return enriched data via ValidationResult.accept_with()
instead of only accepting or rejecting operations. The engine applies
returned fields (contents, tags, tag_groups) to the operation parameters.
- Add accept_with() factory to ValidationResult with optional enrichment
fields: contents, tags, tags_match, tag_groups
- Add tags, tags_match, tag_groups to RecallContext so validators can
see current filter state
- Update _validate_operation to return ValidationResult
- Apply enrichment from result at all retain (2 sites) and recall call
sites in MemoryEngine
- Existing validators using accept()/reject() work unchanged
LLM providers like MiniMax wrap JSON responses in markdown code fences
(```json ... ```), causing JSON parse failures and 5-11 retries per
extraction. The existing fence stripping logic was gated to only
"lmstudio" and "ollama" providers (and for Ollama, unreachable due to
the _call_ollama_native redirect).
Changes:
- Extract _strip_code_fences() helper function
- Apply fence stripping to all providers in call() (not just local)
- Add fence stripping safety net to _call_ollama_native()
- Add 10 tests covering bare JSON, fenced JSON, malformed fences,
and real-world MiniMax response format
Fixesvectorize-io/hindsight#645
Co-authored-by: feniix <feniix@desktop>
* fix(recall): reject empty queries with 400 and fix SQL parameter gap causing IndeterminateDatatypeError
When query text contains only punctuation/symbols (no word characters after
normalization), the BM25 arms are skipped but the old code still placed `limit`
at \$3 in the params list. If tags or tag_groups were also set, their params
(\$4+) were referenced in the SQL while \$3 was a gap, causing PostgreSQL to
raise IndeterminateDatatypeError.
Fix the parameter layout so `limit` is only appended to params when tokens are
present (i.e. when BM25 arms actually use LIMIT \$3), and shift tags_param_idx
from 4 to 3 in the no-tokens path.
Also add a field_validator on RecallRequest.query that rejects empty-after-
normalization queries at the API layer with a 400 before they reach the DB.
* refactor: extract tokenize_query helper and reuse in RecallRequest validator
Remove the sys_platform == 'darwin' constraint that prevented
claude-agent-sdk from installing on Linux, breaking the claude-code
provider in Docker containers.
Fixes#640
* fix: MCP tool calls fail when MCP_AUTH_TOKEN and TENANT_API_KEY differ
When both HINDSIGHT_API_MCP_AUTH_TOKEN and ApiKeyTenantExtension are
configured with different values, MCP transport auth passes but tool
execution fails because the MCP token gets re-validated against the
tenant API key in the engine layer.
Add mcp_authenticated flag to RequestContext so the engine skips tenant
re-validation when MCP transport auth already succeeded.
Fixes#627
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* test: strengthen assertion to verify no auth error in tool response
The original test only checked that "banks" key existed in the response,
which was true even for error responses like {"error": "...", "banks": []}.
Now asserts "error" not in parsed to properly catch auth failures.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat: add fact_types and mental model exclusion filters to reflect and mental models
Adds three new filtering options to both the reflect endpoint and mental model creation/refresh:
- `fact_types`: restrict which fact types (world, experience, observation) are retrieved.
Disables irrelevant agent tools entirely (no wasted tokens).
- `exclude_mental_models`: skip the search_mental_models tool altogether.
- `exclude_mental_model_ids`: exclude specific mental models by ID (merged with the
existing self-exclusion logic during mental model refresh).
For mental models, options are persisted in the existing `trigger` JSONB column so they
are automatically applied on every refresh. The `UpdateMentalModelRequest` already
proxies `trigger`, so no extra endpoint changes are needed.
Also fixes the test fixture (`pg0_db_url` in conftest.py) to correctly resolve pg0://
URLs and run migrations before tests, which was causing all DB-dependent tests to fail
with "relation public.banks does not exist" when HINDSIGHT_API_DATABASE_URL=pg0://uuuu.
* fix: guard against disabled-tool hallucination and regenerate clients
- Add enabled_tools guard in reflect agent: if an LLM calls a tool that
was excluded (e.g. recall when fact_types=["observation"]), return an
error result instead of executing it
- Regenerate OpenAPI spec and all SDK clients (Go, Python, TypeScript)
to include new fact_types / exclude_mental_models fields
* fix: add missing ReflectRequest fields in Rust CLI struct initializers
* fix: filter hallucinated tool calls before trace to prevent disabled tools appearing in results
* chore: merge main, fix lint formatting and update skills openapi.json
* feat: expose fact_types, exclude_mental_models, exclude_mental_model_ids in control plane UI
* fix: add missing trigger fields to MentalModel type in control plane api.ts
* fix: add missing trigger fields to local MentalModel interface in mental-models-view
* feat: tabbed mental model dialogs (Basic / Options tabs)
* refactor: shared FactTypeFilter component, tabbed mental model dialogs use General tab, clean up labels
* feat: pill-style toggle buttons for fact type filter (blue/emerald/amber per type)
* fix: add spacing between Fact Types label and pills, rename to Exclude all mental models
* Fix non-atomic async operation creation in _submit_async_operation
Previously the method performed two separate database round-trips:
1. INSERT into async_operations with no task_payload (null)
2. submit_task → UPDATE to set task_payload
A process crash or network error between steps 1 and 2 left a row with
task_payload IS NULL permanently. The worker's claim query requires
task_payload IS NOT NULL, so these orphaned rows could never be picked up
and the queue appeared degraded indefinitely.
Fix: build full_payload before the INSERT and include task_payload in the
same INSERT statement, making operation creation atomic. submit_task is
still called afterwards — for SyncTaskBackend it executes the task
immediately (unchanged behaviour); for BrokerTaskBackend it becomes an
idempotent UPDATE (payload already set) kept for symmetry.
* Preserve datetime payloads in atomic async insert
* Fix orphaned batch_retain parents when child fails via unhandled exception
When a child retain operation fails with an unhandled exception (e.g. a DB
constraint violation), the memory engine's transaction is rolled back entirely,
including any call to _maybe_update_parent_operation. The poller's fallback
_mark_failed then updates the child status but leaves the parent batch_retain
permanently stuck in 'pending'.
Fix: wrap _mark_failed in a transaction and call a new poller-level
_maybe_update_parent_operation after marking the child failed. This mirrors
the memory engine's own parent-update logic and ensures the parent is
resolved to completed/failed regardless of how the child failure was detected.
The poller's implementation locks the parent row, checks all siblings, and
only finalises the parent once all siblings have reached a terminal state.
Errors in parent propagation are logged but do not affect the child failure
path, which is the critical state change.
* Add tests for _mark_failed parent propagation in WorkerPoller
Tests cover the new _maybe_update_parent_operation logic:
- Last sibling fails → parent batch_retain becomes failed
- Sole child fails → parent becomes failed
- Sibling still pending → parent stays pending (no premature resolution)
- No parent in result_metadata → safe no-op
- End-to-end: unhandled exception via execute_task propagates to parent
* Fix entity_id null constraint for non-ASCII entity names (Turkish İ etc.)
Python's str.lower() and PostgreSQL's LOWER() produce different results for
some Unicode characters. The most common case is Turkish İ (U+0130):
Python: 'İstanbul'.lower() == 'i\u0307stanbul' (i + combining dot, 2 chars)
PostgreSQL: LOWER('İstanbul') == 'istanbul' (plain i, 1 char)
In _resolve_from_candidates, the fallback SELECT for conflicted entity names
passed Python-lowercased strings to LOWER(canonical_name) = ANY($names), so
PostgreSQL couldn't match them. entity_ids[idx] stayed None, which then
caused a NOT NULL violation on unit_entities.entity_id, failing the entire
retain.
Fix: pass original mixed-case names to the fallback SELECT and use
LOWER(canonical_name) = ANY(SELECT LOWER(n) FROM unnest($2) AS n) so
PostgreSQL lowercases both sides identically. The query also returns the
original input_name so we can add a Python-lowercased key to id_by_name
for the assignment loop that uses Python-lowercased keys.
* Add regression test for Unicode entity conflict
The Pydantic model extraction paths (batch API and parallel extraction) used
fact_from_llm.fact_type directly, bypassing the \"assistant\" → \"experience\"
conversion and causing DB CHECK constraint violations.
Unified the conversion logic across all paths:
- \"assistant\" → \"experience\"
- \"world\" → \"world\"
- anything else: fall back to fact_kind (\"assistant\" → \"experience\"), else \"world\"
* feat: upgrade MiniMax default model from M2.5 to M2.7
MiniMax has released MiniMax-M2.7, their latest model with a 1M context
window (up from 204K). This updates the default model across config,
docs, and examples. M2.5 remains fully compatible for users who prefer it.
- Update PROVIDER_DEFAULT_MODELS to MiniMax-M2.7
- Update .env.example and documentation references
- Add test_minimax_provider.py with M2.7 and backward compat tests
* chore: remove test file per review feedback
---------
Co-authored-by: PR Bot <pr-bot@minimaxi.com>
* fix: prevent silent memory loss on consolidation LLM failure
When all LLM retries are exhausted during consolidation, memories were
being marked consolidated_at unconditionally, permanently excluding them
from future consolidation runs without producing any observations.
Fix with two complementary mechanisms:
- Adaptive batch splitting: on LLM failure, the batch is halved and
retried recursively down to batch_size=1, recovering most transient
failures (rate limits, Pydantic validation on long prompts) without
operator intervention
- consolidation_failed_at column: only single-memory batches that still
fail after all retries are marked here instead of consolidated_at, so
they remain visible and retryable
- New API endpoint POST /v1/default/banks/{bank_id}/consolidation/retry-failed
resets these memories for the next consolidation run
* chore: regenerate OpenAPI spec
* fix: rename consolidation endpoint from /retry-failed to /recover
* fix: add consolidation_failed_at column, adaptive batch splitting, and recovery API
- Migration a3b4c5d6e7f8: add consolidation_failed_at TIMESTAMPTZ column to
memory_units with an index for efficient failure queries; properly chains off
g7h8i9j0k1l2 (backsweep_orphan_observations)
- Consolidator: filter pending memories with consolidation_failed_at IS NULL
so failed memories are not re-fetched in an infinite loop
- Consolidator: adaptive batch splitting — when a batch exhausts all 3 LLM
retries, halve it and retry sub-batches recursively; only single-memory
batches that also exhaust all retries get consolidation_failed_at set
- New tests (9 total) covering: adaptive splitting recovers all memories,
larger batch splitting, single-memory permanent failure, exclusion from
next run, partial batch failure, recover resets columns, recover returns
0 when none failed, recover-then-consolidate succeeds, HTTP endpoint
* chore: regenerate Go, Python, TypeScript clients with recover consolidation endpoint
* feat: add Recover Consolidation action to bank Actions dropdown
* style: apply ruff formatting to http.py and config.py
* fix: handle consolidation scope in large batch test mock LLM
The mock LLM was returning {"facts": ...} for ALL calls including consolidation.
Consolidation doesn't use skip_validation=True so it expects a _ConsolidationBatchResponse
instance, not a raw dict. Before this PR consolidation silently swallowed the AttributeError
(failed=False was returned); now failed=True triggers adaptive splitting and timeouts.
Fix: return _ConsolidationBatchResponse() when scope=="consolidation".
* fix: restrict claude-agent-sdk to macOS platform only (no Linux wheel available)
Also fix pre-existing type errors: use setattr for XLM-RoBERTa monkey-patch
and add missing reranker_local_fp16/bucket_batching/batch_size fields to main.py config constructor.
* fix: add UV_INDEX_STRATEGY=unsafe-best-match to fix markupsafe cp314 wheel conflict
PyTorch CPU index serves markupsafe==3.0.3 with only cp314 wheels.
uv's default first-index strategy stops at the first index with any version
even if no compatible wheel exists. unsafe-best-match searches all indices
for the best compatible wheel, falling back to PyPI for markupsafe.
* fix: use explicit pytorch index to prevent markupsafe wheel conflict
Configure the pytorch CPU index as explicit=true in pyproject.toml so it is
ONLY used for torch (via [tool.uv.sources]). All other packages (including
markupsafe) are resolved exclusively from PyPI, preventing the pytorch index
from serving incompatible cp314-only wheels for non-pytorch packages.
Remove UV_INDEX and UV_INDEX_STRATEGY from CI workflow (no longer needed
since the index is now configured in pyproject.toml).
* ci: trigger CI run
* ci: retry trigger
* ci: trigger after remote URL fix
* ci: add workflow_dispatch to unblock manual trigger
* fix: remove empty env blocks left after UV_INDEX removal
* fix: add type: ignore for optional claude_agent_sdk imports (macOS-only)
* fix: correct type: ignore rules for claude_agent_sdk and fix utcnow deprecation
* feat(retain): add verbatim extraction mode
Adds retain_extraction_mode="verbatim" that stores each chunk as-is
without LLM summarization. The LLM still runs to extract entities,
temporal info, and location for full indexability — only the fact text
is replaced with the original chunk content (one memory per chunk).
Useful for RAG-style indexing and benchmarks where original text
must be preserved in memory.
- Add "verbatim" to RETAIN_EXTRACTION_MODES in config.py
- Add VERBATIM_FACT_EXTRACTION_PROMPT with instructions to preserve text
- Add _collapse_to_verbatim() post-processing to enforce 1 fact/chunk
- Expose in bank config UI dropdown with updated description
- Update configuration.md docs with verbatim mode description
- Add unit test for _collapse_to_verbatim and integration test via LLM
- Fix pre-existing main.py CLI override missing new reranker fields
- Fix pre-existing cross_encoder.py ty type error via setattr
* refactor(retain): verbatim mode skips 'what' field entirely
Instead of asking the LLM to echo the chunk text back into 'what' and
then discarding it, verbatim mode now uses a dedicated schema
(VerbatimExtractedFact) that omits the 'what' field altogether.
The LLM only returns metadata (entities, temporal info, location, who),
saving output tokens and avoiding any risk of paraphrasing before the
backfill.
- Add VerbatimExtractedFact / VerbatimFactExtractionResponse models
- Verbatim mode skips causal-relations section (nothing to relate causally)
- _extract_facts_from_chunk: allow missing 'what' in verbatim mode,
set combined_text="" (backfilled by _collapse_to_verbatim)
- Update verbatim prompt to say DO NOT include 'what'
* feat(retain): add index_only extraction mode
Zero-LLM retain mode: chunks are stored as-is with no LLM call, no
entity extraction, and no temporal indexing. Embeddings still run for
semantic search. User-provided entities via RetainContent.entities
are the sole source of entity data.
Early return placed before the batch-API check so no LLM queue or
concurrency locks are acquired.
- Add "index_only" to RETAIN_EXTRACTION_MODES
- Add _extract_facts_index_only() with pure Python chunking path
- Add to UI dropdown and update description
- Update configuration.md with index_only docs and table entry
- Add unit test asserting zero token usage and exact text preservation
* feat(retain): add named retain strategies
Allows mixing extraction modes in a single bank via named strategies.
Each strategy is a set of hierarchical config overrides (extraction_mode,
chunk_size, entity_labels, entities_allow_free_form, etc.) applied on
top of the resolved bank config at retain time.
- retain_strategies: dict of strategy_name → config overrides (bank config)
- retain_default_strategy: default strategy when none specified (bank config)
- strategy field on /retain request: per-call override
- apply_strategy() in config_resolver applies overrides via dataclasses.replace()
- strategy propagates through retain_batch_async → _retain_batch_async_internal
and through the async worker task payload
- Any hierarchical field is overridable per strategy, including entity_labels
and entities_allow_free_form
- Docs updated with strategy configuration example and RRF fairness note
- Unit test for apply_strategy covering overrides, unknown strategy, and
non-hierarchical field filtering
* feat(retain): add per-item strategy and strategy tests
- Add `strategy` field to `MemoryItem` so individual items in a retain
request can override the request-level strategy
- Add `strategy` field to `FileRetainMetadata` for per-file strategy
override in file retain requests
- Group memory items by effective strategy in `api_retain`; each group
is processed as a separate batch, results are aggregated
- Thread strategy through `submit_async_file_retain` →
`_handle_file_convert_retain` → retain task payload
- Add `operation_ids` to `RetainResponse` for async requests with
mixed per-item strategies
- Add `test_strategy_overrides_extraction_mode_for_index_only`: unit
test verifying a named strategy with index_only bypasses the LLM
- Add `test_retain_request_per_item_strategy_field`: unit test for
per-item strategy grouping logic
* feat(ui): add retain strategies and default strategy to bank config UI
- Add StrategiesEditor component: per-strategy cards with name input and
JSON overrides textarea; supports add/remove; validates JSON inline
- Add Default Strategy text input (retain_default_strategy)
- Update RetainEdits type and retainSlice() to include both new fields
- Regenerate OpenAPI spec (retain_strategies, retain_default_strategy,
per-item strategy on MemoryItem/FileRetainMetadata, operation_ids on
RetainResponse)
* refactor(ui): move retain strategies into its own dedicated config section
* feat(ui): improve retain strategies UX and add strategy to document dialog
- Strategy form now includes entity section (free form toggle + entity labels editor)
- Default strategy selector moved outside tab panel, above strategy chips
- Strategy tabs redesigned with underline indicator style for clarity
- Remove strategy confirms with AlertDialog
- Fix tab re-render bug when typing strategy name (skipSyncRef)
- Add strategy field to Add New Document dialog (text + per-file for uploads)
- File upload collapsible uses same Document/Tags/Source tabbed layout
- API: validate empty strategy names in config_resolver
- api.ts: add strategy field to retain and uploadFiles types
* fix: forward strategy through HTTP layer and SDK; add integration test
- route.ts: extract and forward `strategy` from request body to retainBatch
- TypeScript SDK: accept and forward `strategy` in retainBatch options and per-item
- config_resolver.py: validate empty strategy name keys on update
- bank-config-view.tsx: merge entity fields into RetainStrategyForm, redesign strategy tabs with underline style, add confirmation dialog for removal, fix tab-reset-on-typing with skipSyncRef, move default strategy selector outside panel
- bank-selector.tsx: add strategy field to Add Document dialog (per-file in tabbed collapsible)
- test_retain.py: add end-to-end integration test verifying named strategy application (index_only = 0 LLM tokens)
* fix: regenerate TypeScript client with strategy field in RetainRequest/MemoryItem
- Regenerate OpenAPI spec to include strategy field in RetainRequest and MemoryItem
- Regenerate TypeScript client from updated spec
- Add strategy to MemoryItemInput interface
- Remove (item as any) cast now that strategy is properly typed
* rename: index_only extraction mode → chunks
* remove top-level strategy from RetainRequest; strategy is per-item only
* fix(clients): update Go and Python generated clients with strategy/operation_ids fields
* fix(ci): update hierarchical field count, add strategy to Rust MemoryItem initializers
* fix(go-client): minimal targeted YAML updates for strategy/operation_ids fields
