engram
A bitemporal, graph-backed memory system for AI coding agents.
Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.
How we verified
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
rawcontext
Source: github_code
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →
Install from
The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.
These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.
contextAssemble comprehensive context for a task by combining: semantic memory search, past decisions, and file modification history. Use PROACTIVELY at the START of complex tasks to prime yourself with institutional knowledge before diving in. More thorough than recall alone - automatically searches multi…
enrich_memoryAuto-generate metadata for memory content before storing. Returns: one-line summary, searchable keywords, and suggested category (maps to memory type). Recommended workflow: call enrich_memory first, then pass the enriched metadata to remember for better future retrieval. Requires client sampling ca…
extract_factsParse unstructured text into discrete, atomic facts suitable for storage. Use before remember when processing: documentation, chat logs, meeting notes, or verbose command outputs. Each extracted fact can be stored and searched independently, improving retrieval precision. Requires client sampling ca…
queryQuery the knowledge graph directly using Cypher for complex lookups that semantic search cannot handle. Use when you need to: find all decisions within a date range, trace relationships between sessions and files, count memories by type, or explore graph structure. Only read operations are allowed.
recallSearch past memories using semantic similarity and knowledge graph traversal. Use PROACTIVELY: at session start to prime yourself with relevant prior knowledge, before making decisions to check for existing rationale, or when the user references 'before', 'last time', or 'remember when'. Returns mem…
rememberPersist valuable information to long-term memory for future sessions. Use PROACTIVELY when you learn: user preferences, architectural decisions, project conventions, debugging insights, or facts worth preserving. Memories are searchable across sessions and survive context boundaries.
summarizeCondense long text into key points using the MCP client's LLM. Use before storing memories to create compact, searchable summaries. Also useful for: distilling verbose error logs, compressing context that exceeds limits, or creating session recaps. Requires client sampling capability - returns avail…
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
@modelcontextprotocol/sdk has cross-client data leak via shared server/transport instance reuse
Anthropic's MCP TypeScript SDK has a ReDoS vulnerability
AUTH_DATABASE_URLAUTH_ENABLEDENGRAM_API_URLENGRAM_AUTH_SERVER_URLENGRAM_AUTH_TOKENENGRAM_FORCE_OAUTHENGRAM_INGESTION_URLENGRAM_MCP_CLIENT_IDENGRAM_MCP_CLIENT_SECRETENGRAM_MCP_SERVER_URLENGRAM_OBSERVATORY_URLENTITY_DETECT_RELATIONSHIPSENTITY_EXTRACTION_ENABLEDENTITY_MAX_PER_MEMORYENTITY_RESOLUTION_THRESHOLDFALKOR_URLGEMINI_API_KEYGRAPH_RETRIEVAL_ENABLEDGRAPH_RETRIEVAL_MAX_DEPTHGRAPH_RETRIEVAL_WEIGHTLOG_LEVELMAX_SESSIONS_PER_USERMCP_HTTP_PORTMCP_TRANSPORTSEARCH_URLSESSION_TTL_SECONDSTool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
7/7 tools missing one or more hints — context (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); enrich_memory (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); extract_facts (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +4 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.
For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.
Tool handlers catch errors
Only 3/7 tool handlers wrap calls in try/catch (43%)
Wrap each tool handler body in try/catch and return a structured error response.
License file
No license file
Add a LICENSE file (MIT, Apache-2.0, etc.).
Tool test coverage
4/7 tools referenced in tests (57%)
Write tests that reference each tool by name so every tool has at least one test.
Production dependencies are patched
0 critical, 2 high severity in production deps — @modelcontextprotocol/sdk@1.25.1 (high), @modelcontextprotocol/sdk@1.25.1 (high)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.
[](https://m8ven.ai/mcp/rawcontext/engram)?variant=verified from the URL.Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.
https://m8ven.ai/api/mcp/tool-check