personal-kg-mcp (tomschell/personal-kg-mcp) is an MCP server listed on the M8ven Trust Index. It scores 39 out of 100, grade F. It declares 18 tools. No publisher has claimed this listing.
Auto-captures decision context from multi-agent workflows to preserve the 'why' behind every choice. Enables task traceability, reasoning retrieval, and continuous improvement across planning and implementation sessions.
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
tomschell
Source: Glama
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.
kg_analyzeUnified analysis tool. Supports: 'clusters' for topic grouping, 'emerging' for trend detection, 'graph_export' for full export, 'path' for finding connections between nodes.
kg_capturePrimary tool for capturing knowledge nodes. Use this to record decisions, progress updates, insights, questions, and ideas. Automatically creates relationships, normalizes tags, and links to sessions. This is your main entry point for adding knowledge to the graph.
kg_link_sessionCreates a relationship between a session node and another knowledge node. Use to explicitly link work items, decisions, or progress to a specific session for better organization.
kg_update_nodeUpdates an existing knowledge node. Use to modify content, tags, importance, or visibility of a node. Supports partial updates - only specified fields are changed.
kg_diagnosticGenerate a diagnostic HTML dashboard for debugging knowledge graph context selection, search results, and node visibility. Opens in browser to help understand why certain nodes are included/excluded from context.
kg_adminUnified admin tool for maintenance operations. Supports: 'health' for system status, 'backup' for data protection, 'validate' for integrity checks, 'repair' for fixing issues, 'export'/'import' for data migration, 'rename_tag'/'merge_tags' for tag management, 'migrate_embeddings' for backfilling Ope…
kg_metricsUnified metrics tool. Supports: 'snapshot' for period overview, 'daily' for specific date, 'timeseries' for trends, 'tool_calls' for tool usage, 'searches' for search patterns, 'cleanup' for maintenance.
kg_get_project_stateProvides a comprehensive overview of a project's current state including active focus areas, recent decisions, open questions, blockers, and completed tasks. Perfect for project status checks and planning.
kg_session_warmupStart every session with this tool! Loads comprehensive context about your project including recent work, active questions, and blockers. Essential for maintaining continuity between work sessions.
kg_nodeUnified tool for node operations. Supports three operations: 'get' to retrieve a node with its relationships, 'delete' to remove a node, 'find_similar' to find semantically similar nodes.
kg_capture_sessionCaptures session summaries with structured metadata. Use at the end of work sessions to record what was accomplished, artifacts created, and next actions. Essential for maintaining context between sessions and tracking progress over time.
kg_open_questionsLists unresolved questions with aging information. Questions are considered 'stale' after 3 days. Use to track what needs follow-up and identify forgotten questions.
kg_resolve_questionMarks a question as resolved by linking it to a decision, insight, or other node that answers it. Creates a 'resolved_by' relationship.
kg_edgesUnified tool for relationship operations. Supports: 'create' to link nodes, 'list' to view relationships, 'maintain' for cleanup and rebuilding.
kg_searchUnified search tool for finding knowledge nodes. Supports three modes: 'text' for keyword/tag filtering with ranking, 'semantic' for AI-powered meaning-based search, 'time_range' for date-based queries. Default mode is 'semantic' for best results.
kg_query_contextReconstructs context around a specific topic by analyzing related knowledge nodes. Use to understand the full context and background of a particular subject area.
kg_list_tagsLists all tags in the knowledge graph with usage counts. Use to discover available tags, find inconsistencies, or identify commonly used categories.
kg_get_relevant_contextRetrieves relevant past context for a given query. Returns brief summaries of related decisions, insights, and open questions. Perfect for proactive context injection before starting work on a topic.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
When Vitest UI server is listening, arbitrary file can be read and executed
@modelcontextprotocol/sdk has cross-client data leak via shared server/transport instance reuse
Anthropic's MCP TypeScript SDK has a ReDoS vulnerability
Model Context Protocol (MCP) TypeScript SDK does not enable DNS rebinding protection by default
OPENAI_API_KEYOpenAI API key for semantic embeddingsPKG_EMBEDDING_DIMPKG_EMBEDDING_MODELOpenAI embedding modelPKG_EMBEDDING_STARTUPPKG_GITHUB_INTEGRATION_ENABLEDEnable GitHub issue/PR integrationPKG_GITHUB_TOKENAdd to .env file: =github_pat_your_token_herePKG_MCP_CAPTURE_AUTOAuto-capture without explicit callsPKG_MCP_CAPTURE_ENABLEDAuto-capture MCP tool callsPKG_MCP_CAPTURE_EXCLUDETools to exclude (comma-separated)PKG_MCP_CAPTURE_TOOLSTools to capture (comma-separated)PKG_METRICS_ENABLEDPKG_METRICS_RETENTION_DAYSPKG_STORAGE_DIRDirectory for storing knowledge graph dataPKG_USE_ANNUse approximate nearest neighbor searchTool 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
18/18 tools missing one or more hints — kg_analyze (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); kg_capture (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); kg_link_session (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +15 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.
Descriptions match behaviour
1 tool describes read intent but its handler mutates — kg_diagnostic (line 1025: fs.mkdirSync(dir, { recursive: true }))
Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.
Shell command execution
6 calls in production code run through a shell (src/utils/github.ts:58, src/utils/github.ts:59, src/utils/github.ts:86)
Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.
Production dependencies are patched
0 critical, 3 high severity in production deps — @modelcontextprotocol/sdk@1.17.3 (high), @modelcontextprotocol/sdk@1.17.3 (high)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
Dev dependencies
1 critical/high in dev-only deps (does not ship to users)
Upgrade dev dependencies when convenient.
Tool description accuracy
kg_diagnostic: description implies read-only but handler writes/deletes/executes
Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.
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/tomschell/personal-kg-mcp)?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