claude-runway (Donelle/claude-runway) is an MCP server listed on the M8ven Trust Index. It scores 74 out of 100, grade C. It declares 16 tools. No publisher has claimed this listing.

Back to MCP Trust IndexM8ven
C
Emerging
74/100
11 days ago

claude-runway

MCP servers that offload codebase search and log/output compression to local compute (Qdrant + LM Studio), cutting Claude Code's token usage

→ Donelle/claude-runway· listed on github_code

Emerging. No concerning findings. Grades remain capped until the project builds reputation through adoption. 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

Donelle

Source: github_code

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 →

⚡ Connect GitHub for Live Monitored

Install from

GitHub

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
✅
No credential exfiltration, no sensitive file access, no obfuscation
Static analysis found nothing flowing your secrets to unexpected places.
✅
Open source with a license and README
Anyone can audit the code, the license is declared, and the publisher documents what it does.
🔐
You'll be asked for 1 credential: QDRANT_API_KEY
These are read from process.env at runtime. Make sure you trust where they’ll be sent.
// 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.
configQDRANT_URL
🔐 secretQDRANT_API_KEYQDRANT_URL=http://localhost:6333; leave empty for this local
configCOMPACT_COLLECTION
configCOLLECTION_NAME
configEMBEDDING_MODELChanging after a collection has data requires a full index_repo --reset rebuild — old and new vectors aren't compatible.
configCOLLECTION_DESCRIPTION
configINDEX_INCLUDE_EXTENSIONScomma-separated extensions (e.g. ".py,.md") — overrides the built-in list entirely, so only these get ingested.
configINDEX_EXCLUDE_DIRScomma-separated folder names (e.g. "fixtures,generated") — adds to (not replaces) the built-in excludes.
configTOOLS_REPO_DIR
configCLAUDE_RUNWAY_COMPRESS_THRESHOLD_CHARSshell only — the hook is the sole reader, so an .mcp.json copy is inert
configHOOK_DEBUG_LOG
configCLAUDE_RUNWAY_LMSTUDIO_URLCLI: python tools/doctor.py /path/to/target-repo reports any mismatch between .mcp.json's local-compress env block and the live shell for CLAUDE_RUNWAY_LMSTUDIO_URL/_MODEL/_TRACK_SAVINGS/_SAVINGS_DB instead of leaving that silently split ([#49](https://github.com/CVNA-SandboxOrg/claude-runway/issues/49)). Exit code 1 on a mismatch, so it doubles as a CI/pre-flight gate.
configCLAUDE_RUNWAY_CACHE_DB.mcp.json only, if overridden — unlike every other var above, no hook reads this one, so there's no "shell too" half to this rule; the default needs no coordination at all. Safe to delete this file anytime — everything in it is a disposable, always-rebuildable cache, deliberately kept separate from CLAUDE_RUNWAY_SAVINGS_DB's file, which the user wants kept forever
configCLAUDE_RUNWAY_LMSTUDIO_MODELboth
configCLAUDE_RUNWAY_TRACK_SAVINGSSessionEnd hook for the opt-in savings tracker — rolls the session's transient ledger into the perpetual SQLite store and surfaces a short summary via systemMessage. Silent no-op unless CLAUDE_RUNWAY_TRACK_SAVINGS is on and the session logged at least one credited event. See "Savings tracker" below.
configCLAUDE_RUNWAY_SAVINGS_DBboth if overridden — the default needs no coordination, since every script derives it identically
// 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

16/16 tools missing one or more hints — list_local_models (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); compress_file (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); compress_command_output (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +13 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.

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 2 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 Score](https://m8ven.ai/badge/mcp/donelle/claude-runway)](https://m8ven.ai/mcp/donelle/claude-runway)
Shows your grade and updates automatically. Prefer no grade? Append ?variant=verified to the badge URL.
commit: 310c37c5e5d8bca7ede41f7ac9e0a5a553c188f1
code hash: 40ab4faa4458d5eea28b2bbcff1ce339e1a8087609314f8eedec13d12b435d61
verified: 9/2/2026, 6:50:04 PM
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