Cordelia Proxy (seed-drill/cordelia-proxy) is an MCP server listed on the M8ven Trust Index. It scores 26 out of 100, grade F. It declares 21 tools. No publisher has claimed this listing.

F
Warning
26/100

Cordelia Proxy

MCP server for memory storage, encryption, search, and group sharing, used by Claude Code via the cordelia-agent-sdk.

Warning. Serious findings were identified. Review the full report before connecting. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.

How we verified

Code Verified⚡ Live Monitored: not connected

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

seed-drill

Source: Glama

Is this your MCP?

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.

// key findings
🚨
Secret credentials may flow to a network call
7 flows detected: CORDELIA_NODE_TOKEN, CORDELIA_API_KEY. We can’t prove the destination matches the brand the credential belongs to.
⚠️
Known vulnerabilities in dependencies: 2 high
Affects packages this MCP installs at runtime. Upgrade or remove the affected dependency.
🔐
You'll be asked for 5 credentials: CORDELIA_API_KEY, CORDELIA_NODE_TOKEN, CORDELIA_SESSION_SECRET, GITHUB_CLIENT_SECRET, HONEYCOMB_API_KEY
These are read from process.env at runtime. Make sure you trust where they’ll be sent.
// tools this server exposes21 tools

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.

memory_read_hot

Read L1 hot context for a user. Returns dense structured memory including identity, active state, preferences, and delegation rules. Load this at session start.

memory_write_hot

Write to L1 hot context for a user. Use patch for partial updates, replace for full replacement. Supports optimistic concurrency via expected_updated_at.

memory_status

Get memory system status and available users.

memory_analyze_novelty

Analyze text for novelty signals. Returns detected signals, confidence scores, and suggested persistence targets. Use this to determine what from a conversation should be persisted to memory.

memory_search

Search L2 warm index by keyword, type, and/or tags. Returns matching entries with relevance scores. Use this to find entities, sessions, or learnings from memory.

memory_read_warm

Read a specific L2 warm item by ID. Returns the full item content (entity, session, or learning).

memory_write_warm

Create or update an L2 warm item. Use for storing entities (people, orgs, projects, concepts), session summaries, or learnings (patterns, insights, principles).

memory_delete_warm

Delete a specific L2 warm item by ID. Removes the item from storage and the index.

memory_backup

Create a backup of the memory database. Returns backup manifest with SHA-256 verification. SQLite provider only.

memory_restore

Restore memory database from a backup. Verifies SHA-256, runs integrity checks, migrates schema if needed. SQLite provider only.

memory_share

Share a private memory to a group. Creates an immutable copy; the original is never modified.

memory_group_create

Create a new group. The calling entity becomes owner.

memory_group_list

List groups, optionally filtered by entity membership.

memory_group_read

Read group details including members.

memory_group_add_member

Add a member to a group. Requires admin/owner role.

memory_group_remove_member

Remove a member from a group. Requires admin/owner role.

memory_bind_context

Bind a working directory to a group. Writes from that directory will be scoped to the group, and searches will prioritize group memories. Stored in L1 active.context_bindings.

memory_unbind_context

Remove a context binding for a working directory. Restores default behavior (private memory).

memory_backfill_embeddings

Rebuild FTS and vec indexes from L2 items. Rebuilds FTS keyword index (includes details field) and vec embeddings. Uses cached embeddings where available, generates new ones via Ollama.

memory_reconcile

Reconcile search index with node storage. Indexes replicated items that arrived via P2P but are missing from local FTS5/vec search. Runs automatically on startup and every 60s, but can be triggered manually.

memory_prefetch_l2

Prefetch top L2 items for faster session start. Returns most recently accessed items from user groups and private memory. Context-aware: prioritizes bound group if context binding exists.

// known CVEs in dependencies2 high

Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.

high@opentelemetry/auto-instrumentations-node@0.69.0GHSA-q7rr-3cgh-j5r3

Prometheus exporter process crash via malformed HTTP request

high@opentelemetry/sdk-node@0.211.0GHSA-q7rr-3cgh-j5r3

Prometheus exporter process crash via malformed HTTP request

Depend on this server? Get alerted when its CVEs change.Watch this server free →
// environment variables
To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.
🔐 secretCORDELIA_API_KEY
configCORDELIA_BASE_URL
configCORDELIA_CONTEXT_WINDOW
configCORDELIA_CORE_API
configCORDELIA_CREDENTIALS
configCORDELIA_CREDENTIALS_FILE
configCORDELIA_EMBEDDING_MODELnomic-embed-text
configCORDELIA_EMBEDDING_PROVIDERollama
configCORDELIA_EMBEDDING_URLhttp://localhost:11434
configCORDELIA_HOME
configCORDELIA_HTTP_HOST
configCORDELIA_HTTP_PORT
configCORDELIA_INTEGRITY_INTERVAL_MS
configCORDELIA_LOCAL_USERS
configCORDELIA_MEMORY_ROOT
🔐 secretCORDELIA_NODE_TOKEN
configCORDELIA_NODE_URL
configCORDELIA_REMOTE_URL
🔐 secretCORDELIA_SESSION_SECRET
configCORDELIA_STORAGEsqlite
configCORDELIA_TTL_SWEEP_INTERVAL_MS3600000
configFLY_APP_NAME
configFLY_REGION
configGITHUB_CLIENT_ID
🔐 secretGITHUB_CLIENT_SECRET
🔐 secretHONEYCOMB_API_KEY
configHONEYCOMB_DATASET
configHOST
configOTEL_SERVICE_NAME
configPORTAL_URL
Deployment configuration, supplied by whoever hosts the server. Users are not asked for these.
deploySENTRY_DSN
// quality suggestions

Tool 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

21/21 tools missing one or more hints — memory_read_hot (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); memory_write_hot (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); memory_status (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +18 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 test coverage

Only 0/21 tools referenced in tests (0%)

Write tests that reference each tool by name so every tool has at least one test.

Shell command execution

1 call in production code run through a shell (hooks/recovery.mjs:19)

Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.

Secrets not logged

3 secret values sent to error/log

Redact or omit secret values from log output.

Production dependencies are patched

0 critical, 2 high severity in production deps — @opentelemetry/auto-instrumentations-node@0.69.0 (high), @opentelemetry/sdk-node@0.211.0 (high)

Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.

Dependency freshness

1/18 production deps stale: swagger-ui-express@2024-05-31 (2.3y)

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.

// full audit trail
The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.
// improvement guidance — verified publishers only
We have 6 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.
// embed badge in your README
[![M8ven Verified](https://m8ven.ai/badge/mcp/seed-drill/cordelia-proxy?variant=verified)](https://m8ven.ai/mcp/seed-drill/cordelia-proxy)
Shows verification status without the grade. Want the grade badge instead? Remove ?variant=verified from the URL.
commit: 225cef359d0bab40a60d109fc4eff906c2ca8843
code hash: 52b34e0e17b0c0de89d28e435fcf0ccfe2016e7a7c77716e52443210f9965e51
view raw JSON →
Check MCPs from inside your assistant
Tool Check · MCP

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
check_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern
How to add it →Free · no account needed · works in any MCP client