* feat(context): add ErasureCoordinator for cross-store entity erasure
purge_node() is graph-scope by design (#957), so an entity removed from the
graph can survive verbatim as an AgentMemory item and as an embedding. The
changelog names GDPR Article 17 as purge's motivation, and an Article 17
erasure the vector store can still answer queries from is not an erasure --
it is worse than none, because purge_node() returns True and writes a
tombstone attesting the content is gone.
ErasureCoordinator composes the existing public APIs to drive the cascade and
returns an ErasureReceipt recording what each store reported. Nothing in
context_graph.py or agent_memory.py changes behaviorally; ContextGraph keeps
its documented graph-scope contract instead of acquiring references that would
invert the dependency.
Honest partial reporting is the point. Stores report erased / not_found /
not_configured / unsupported / failed, and complete is False when any store
reports unsupported or failed. FAISS, Milvus and Weaviate expose no delete at
all, so erasure genuinely cannot be completed on them today -- the receipt
says so rather than reporting a success it did not achieve.
Erasure runs outward-in (vectors, memory, graph). The tombstone is the durable
attestation, so writing it first would let a crash mid-cascade leave a record
claiming more than happened; erasing the graph last leaves a partial failure
recoverable and honest.
The memory sweep pages until dry and re-queries afterwards rather than
trusting one find_by_entity() call, whose limit=10 default silently truncates
the very check a caller uses to decide the erasure is done. Unsupported vector
backends are detected by probing the wrapped backend, since the VectorStore
facade declares delete_vectors() for every backend and only raises
NotImplementedError once called.
27 tests against real ContextGraph/AgentMemory instances, including the
25-items-on-one-entity regression that fails against a naive single-call
sweep. Full tests/context/ suite: 596 passed.
Closes#1018
* docs(context): document ErasureCoordinator in the context API reference
* test(context): exercise ErasureCoordinator against a real VectorStore
The vector-leg tests asserted the three backend shapes the coordinator
expects -- delete_vectors / delete / neither -- against fakes, which is worth
exactly as much as the assumption that a real store looks like one of them.
VectorStore(backend="inmemory") runs without external services, so it can
hold that assumption to account.
Adds the end-to-end case the receipt actually attests to: a real ContextGraph,
AgentMemory and VectorStore, where the embedding is written by
AgentMemory.store() and has to be gone afterwards. That exercises the memory
leg's own delete_memory() vector cascade rather than the coordinator's model
of it.
The real backend also pins a limit worth knowing before trusting the receipt:
it pops the ids and returns True whether or not they were there, and no
backend offers a portable existence check, so `erased` on the vectors leg
means the store accepted the delete for the ids given -- not that embeddings
were really removed. The memory leg re-queries to confirm and so is the
stronger claim. Documented on STATUS_ERASED, _erase_vectors(), and both status
tables.
tests/context/: 599 passed.
* fix(context): address review findings on ErasureCoordinator
Timestamp drift (high). erase_entity() resolved erased_at up front but passed
the caller's original `at` down to purge_node(), so on the default at=None
path the coordinator and the graph each took their own now() and the receipt
attested to a different instant than the tombstone it points at -- breaking
the invariant this module states most loudly. The resolved value is now what
the graph receives. The existing test passed only because it supplied an
explicit `at`, which hides the drift; the regression test covers at=None,
which is what callers actually use.
Backend delete results. The vectors leg treated anything other than the
literal False as success, but no in-repo backend returns a bool -- Qdrant
returns {"status": <UpdateStatus>} and Pinecone {"deleted": True}, so every
dict read as success and the backend's own account of the delete was thrown
away. Results are now interpreted by shape and the payload is kept in the
receipt as backend_result, stringified so it stays JSON-serializable as an
audit record. Bool markers match by identity so a 0 count isn't read as
False; string markers match as substrings so an enum rendering as
"UpdateStatus.FAILED" isn't read as success.
Falsey vector store. The "at least one store" guard used `not vector_store`,
rejecting a valid store whose __bool__/__len__ makes an empty instance falsey
and then reporting vector_store=None when an object had been passed. It now
separates None (absent) from False (deliberately disabled) from provided, and
echoes what it received.
`at` annotations. Widened to int/float, matching the ContextGraph normalizer
they delegate to, so the coordinator stops advertising less than the API it
wraps.
tests/context/: 608 passed.
* fix(context): report memory-owned vectors that survive erasure (#1018)
A receipt could read complete while an embedding was still in the vector
store. The vector leg deleted `vector_ids` or `[entity_id]`, and the memory
leg relied on `AgentMemory.delete_memory()` to cascade to the vectors each
item owns. That cascade is best-effort: `_delete_vector_ids()` raises when a
backend returns False, `delete_memory()` catches it, logs a warning, and
still returns True. So `batch_delete` counted the item, the residual re-query
found no items, the memory leg reported `erased`, and nothing in the receipt
recorded that the embedding was refused.
Reproduced with a store that deletes the entity-keyed id and refuses the
memory-owned one: `receipt.complete` was True with the embedding still live.
That is the failure mode this module exists to prevent -- a receipt is a
compliance artifact, and one that overstates is worse than none.
Fix by deleting memory-owned vector ids through the coordinator's own vector
leg, which reports honestly, instead of trusting the memory leg's cascade.
The ids are collected before anything is deleted, while the items still exist
to be enumerated, and are unioned with any caller-supplied ids rather than
replacing them.
This needs one addition to AgentMemory: `vector_ids_for(memory_id)`, a
read-only accessor mirroring the fallback in `delete_memory` (an item stored
without tracked ids is keyed by its own memory id). Reaching into
`_vector_ids` from the coordinator would have been the internals-access
pattern this repo keeps getting bitten by. No existing AgentMemory behaviour
changes -- `delete_memory()` still cascades best-effort, so other callers are
unaffected; the coordinator simply no longer depends on that being reliable.
It does mean the vectors are attempted twice, which is a no-op on a working
store and only ever costs a log line.
Note this deviates from the PR's stated "nothing in agent_memory.py changes"
constraint. The constraint could not hold: with `_vector_ids` private and no
portable way to ask a vector store what it still holds, the coordinator had
no way to make the claim truthful without it.
Four tests: the refused-vector case (receipt must be incomplete), that
memory-owned ids reach the store, that explicit `vector_ids` do not displace
them, and the accessor's fallback. The first three were confirmed to fail
against the previous coordinator, on the `receipt.complete` assertion rather
than incidentally. 655 tests pass across tests/context and the agno
integration.
* fix(context): make erasure receipt vector failures honest
* fix(context): optimize erasure pagination handling
* fix(context): use one timestamp for batch erasure
* style: strip trailing whitespace from erasure.py and test file
* docs(changelog): correct test counts to 48 / 738 after review rounds
---------
Co-authored-by: Pravit Ampapathini <pravit.amp@gmail.com>
Co-authored-by: Sameer6305 <sskadam6305@gmail.com>
* docs: improve agent memory guide onboarding and usage guidance
* docs: align Agent Memory guide with persistence implementation
* docs: fix misleading index_path persistence claim across guides
VectorStore's index_path kwarg is silently absorbed into FAISSStore's
**config and never read anywhere in faiss_store.py, so it does not make
the FAISS index persist across restarts as several docs implied. Real
persistence requires an explicit VectorStore.save()/.load() call, or
AgentContext.save()/.load() which cascades to it.
- docs/reference/context.md: rewrite the "Persist your vector store"
tip to explain the actual save()/load() mechanism instead of the
dead index_path kwarg.
- docs/guides/graphrag.md, decision-intelligence.md, ingest.md,
semantic-extraction.md: drop the dead index_path=... kwarg from
VectorStore(backend="faiss", ...) constructor calls.
Follow-up to #691, which fixed the same false claim in
docs/guides/agent-memory.md but missed these other files.
---------
Co-authored-by: KaifAhmad1 <kaifahmad087@gmail.com>
- Fix What's new → link in Info banner (now a proper <a> tag, always clickable)
- Replace 4-stat CardGroup on index with inline premium stats row
- Convert every <CardGroup>/<Card> block site-wide to markdown bullet lists:
content sections → bold-title bullets with sub-bullets, nav cards → [Title](href) — description
- Add cursor-animated list item hover effects to custom.css:
green inset left border, subtle background tint, marker color change on hover
- Affects index, getting-started, quickstart, concepts, modules, faq, architecture,
installation, cookbook, glossary, learning-more, explorer-setup, cli-setup,
community, contributing-guide, governance, citation, project-license,
all integrations pages, and all 20+ reference module pages
- Move Discord, GitHub, PyPI, and Follow on X links from sidebar anchors to top-right navbar
- Lock dark mode as default via appearance.strict and hide theme toggle
- Add custom.css with hover highlighting for tables, code blocks, cards, callouts, and inline code
- Move all Tips and Common Pitfalls sections inline next to their relevant content across all 25 reference docs
- Polish context.md: remove duplicates, condense callouts, upgrade Cookbooks to CardGroup
- Convert Troubleshooting and Performance Optimization sections in installation.md, cli-setup.md, explorer-setup.md, learning-more.md, and faq.md from plain headers to AccordionGroup
- Change navigation-hint Tip callouts to Info in concepts.md, faq.md, glossary.md, and modules.md
* fix: replace invalid Mintlify theme 'venus' with 'mint'
* docs: replace em dashes with colons across all docs files
* fix: strip UTF-8 BOM from all docs files (broke frontmatter detection)
Replace plain markdown in every docs/reference/ file and docs/concepts.md with
rich Mintlify JSX components — CardGroup, Steps, Tabs, AccordionGroup, Tip,
Warning, Note, and CodeGroup — for a consistent, navigable, production-grade
developer experience.
Diagrams (docs/assets/img/diagrams/):
- architecture-overview.svg: 4-column layered architecture
- pipeline-flow.svg: 8-step numbered pipeline flow
- kg-structure.svg: entity/relation graph with typed nodes and labeled edges
- graphrag-flow.svg: dual-path retrieval (vector + graph) to LLM to grounded answer
- extraction-pipeline.svg: NER/Relation/Coreference fan-out to Triplet Generator
- agent-context-flow.svg: AgentContext hub with VectorStore and ContextGraph
- reasoning-chain.svg: forward-chaining inference with explanation path
Wordmark logo (light + dark SVG variants):
- Green rounded-square S icon + Semantica text in green
- docs.json updated to use wordmark SVGs for light and dark modes
Pages updated with diagrams:
- index.md, architecture.md, quickstart.md, concepts.md
- reference/kg.md, reference/pipeline.md, reference/semantic_extract.md
- reference/context.md, reference/reasoning.md
Fixes#433
- ContextGraph.add_decision() now accepts keyword arguments (category,
scenario, reasoning, outcome, confidence, entities, decision_maker)
in addition to a Decision object, matching documented behaviour.
Both call forms return the decision ID string.
- Quickstart snippets in README, getting-started.md, and index.md
changed from VectorStore(backend="faiss") to VectorStore(backend="inmemory")
so they work without faiss-cpu installed.
- docs/reference/context.md methods table updated to reflect the dual
signature of add_decision().
- docs/bugs/quickstart_api_mismatch.md added to track the issue.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add 'from datetime import datetime' import in e-commerce examples
- Change 'max_results=5' to 'limit=5' for find_precedents_by_scenario calls
- Fix docs/reference/context.md e-commerce example
- Fix semantica/context/context_usage.md e-commerce example
- Ensure documentation examples are self-contained and copy-paste ready
- Match actual API parameter names for correct behavior
- All 62 tests still passing successfully
- Fix method name conflicts: add_decision -> add_decision_simple, find_precedents -> find_precedents_by_scenario
- Fix Decision ID handling: align tests with Decision model UUID generation behavior
- Fix AgentContext integration: proper handling of context_graph backend in get_causal_chain
- Fix Policy engine: remove invalid auto_generate_id parameter from deserialization
- Fix node type consistency: handle lowercase 'decision' type across all methods
- Fix timestamp handling: proper conversion for string and datetime objects
- Update documentation: correct method names and Decision model usage in examples
- All 62 Context Graph tests passing successfully
- Production ready with comprehensive verification
- Enhanced README.md with strategic emojis for better visual appeal
- Updated context_usage.md with detailed, user-friendly examples
- Improved docs/reference/context.md with accessible language
- Added AgentContext sections with progressive learning approach
- Maintained professional appearance while improving readability
- Consistent documentation across all context module files
- Remove broken link from reference/context.md that was causing CI failure
- Decision tracking functionality is now integrated into the context module
- Fix mkdocs build strict mode warning about missing target file
- Ensure documentation builds successfully in CI pipeline
- Add decision tracking system with DecisionRecorder, DecisionQuery, CausalChainAnalyzer, PolicyEngine
- Implement KG algorithm integration with centrality, community detection, embeddings, path finding
- Add vector store integration with hybrid search and custom similarity weights
- Enhance context graphs with advanced analytics and decision support
- Update documentation with comprehensive context module reference
- Add production examples for banking and healthcare use cases
- Update README to highlight context graph framework capabilities
- Add comprehensive test suite for all new features
- Reduced code examples in all guide pages (getting-started, quickstart, concepts, modules, examples, use-cases, learning-more)
- Added comprehensive cookbook links with descriptions (topics, difficulty, time, use cases)
- Improved structure and organization across all guide pages
- Updated use-cases.md to only include use cases with corresponding cookbooks
- Removed 'Last Updated: 2024' from all documentation files
- Enhanced navigation with better 'Next Steps' sections
- Add semantica.llms module with Groq, OpenAI, HuggingFace, and LiteLLM providers
- Add query_with_reasoning() method for multi-hop reasoning with LLM-generated responses
- Update ContextRetriever and AgentContext with reasoning capabilities
- Add comprehensive documentation for LLM providers and GraphRAG reasoning
- Update README and docs with new features
- Update notebook examples to use new query_with_reasoning() method
- Removed all Pinecone references, adapters, and documentation to align with open-source, self-hosted focus.
- Removed PineconeAdapter and related dependencies.
- Updated VectorStore to enforce supported backends (FAISS, Weaviate, Qdrant, Milvus, InMemory).
- Updated cookbooks (e.g., 13_Vector_Store.ipynb) to use Weaviate/FAISS examples instead of Pinecone.
- Updated core documentation (modules.md, rchitecture.md, etc.) to reflect backend changes.
- Added new tests ( est_pinecone_removal.py, est_vector_store_deepdive.py) to verify removal and validate remaining backends.
- Verified all vector store tests pass.
