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
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
redaphid
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.
searchSearch past AI conversations. Returns TERSE ranked hits — one line each, no full summaries — so you can triage cheaply, then drill in. Each hit carries: - session_id, title, date, score - matched_tier: which rung matched (session | chunk | message) - snippet: the query-highlighted lead of the match…
getSessionGet a session's DIGEST — summary + chunk map. NO raw messages. BREAKING CHANGE: getSession no longer dumps messages. It returns the middle rung of disclosure: the session summary plus a manifest of its chunks, each with a one-line summary and a { start_message_id, end_message_id } range. Read the r…
getSessionTranscriptResolve a session by external ID or title and return its DIGEST. Despite the name, this aliases getSession (digest, not raw transcript) — it just resolves the session differently: exact external_id match first, then ILIKE title search. Use getMessages to read raw messages once you have the session.
getMessagesRead RAW messages, windowed. Two modes: - Browse a window: { sessionId, offset?, limit? } — default limit 30 (never the whole thread). Page with offset. - Read a chunk's region: { startMessageId, endMessageId } — the id range straight off a getSession chunk-manifest entry. sessionId is inferred…
getMessageRead ONE message in full by id, uncapped. The escape hatch for an oversized message that getMessages returned TRUNCATED — reaching its full content requires this deliberate call, so a huge payload can never arrive by accident.
getChunkGet one chunk's full summary by { sessionId, chunkIndex } — the middle step between a getSession manifest line and reading raw messages, when the one-line manifest summary isn't enough to decide. Includes the message-id range to hand to getMessages.
statsGet statistics about your conversation history
healthReport mindmeld's own indexing health so silent degradation is visible. Surfaces three areas; "unhealthy" cues are documented inline in the output: SUMMARY COVERAGE - total sessions; sessions with a non-NULL summary (excluding deleted + 'Warmup') - coverage % = summarized / (summarized + real NULL…
reportUselessSessionFlag a session that pollutes search results. REVERSIBLE — nothing is deleted. Use this when search returns results that are clearly noise — automated runs, monitoring jobs, notification stubs, sentinel results, tool-call spam, repeated boilerplate, or anything that isn't a real conversation. WHAT …
unreportUselessSessionUndo reportUselessSession. Removes the "useless" tag AND takes the session back out of the noise corpus, so it stops dragging similar sessions down. Agents over-flag — that is expected, and it is why this exists. If a session was reported as noise and it turns out to be real, undo it here; there is…
addTagremoveTagRemove tags from a session or a message — the inverse of addTag, and the reason tagging is safe to do freely: nothing about it is permanent. Removing "useless" from a session returns it to normal search results. Pass exactly one of sessionId / messageId. Removing a tag that was not there is report…
writeNoteDisclosed 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
ALLOWED_HOSTSINGEST_SPOOL_CLIENT_IDINGEST_SPOOL_CLIENT_SECRETINGEST_SPOOL_URLMCP_PORTMINDMELD_DEFAULT_EXCLUDED_TAGSPOSTGRES_DBMust match across every machine feeding one indexPOSTGRES_HOSTNon-standard port to avoid clashing with a local PostgresPOSTGRES_PASSWORDMust match across every machine feeding one indexPOSTGRES_PORTNon-standard port to avoid clashing with a local PostgresPOSTGRES_USERMust match across every machine feeding one indexUI_PORTUPSTREAM_URLWSL_DISTRO_NAMETool 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
13/13 tools missing one or more hints — search (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); getSession (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); getSessionTranscript (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +10 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.
License file
No license file
Add a LICENSE file (MIT, Apache-2.0, etc.).
Readable source code
1 file are minified or bundled, which is usually build output rather than concealment
Ship unminified, readable source.
Secrets stay with their owner
3 secrets sent to a request target we could not resolve (INGEST_SPOOL_CLIENT_SECRET → dynamic, INGEST_SPOOL_CLIENT_SECRET → dynamic) — often a configured endpoint, not necessarily third-party
Audit where credentials are sent. A NOTION_TOKEN should only reach api.notion.com — never a third-party host.
No arbitrary install scripts
Has postinstall/preinstall script — runs arbitrary code on npm install
Remove postinstall/preinstall hooks unless they’re essential.
Dev dependencies
1 critical/high in dev-only deps (does not ship to users)
Upgrade dev dependencies when convenient.
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/redaphid/mind-meld)?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