Propose a patch

Evidence

The sandbox observed nothing beyond the static findings; the proposal cites those.

diff --git a/dist/index.js b/dist/index.js
--- a/dist/index.js
+++ b/dist/index.js
@@ -122,1 +122,0 @@
-// The base as quoted in tool results. A proxy credential in the URL's userinfo is dropped.
@@ -225,1 +224,0 @@
-// upstreamHeaders: empty on stdio; on the hosted endpoint, the forward token and the
diff --git a/dist/http.js b/dist/http.js
--- a/dist/http.js
+++ b/dist/http.js
@@ -32,1 +32,0 @@
-// client's address and a forward token core-backend checks (runtime.ts; core-backend
@@ -39,1 +38,0 @@
-import { forwardToken, markHttpEntrypoint, onUpstreamFailure, requestLabels, requestSettled, SYNTHETIC_UA_FAMILY, withClient } from "./runtime.js";
@@ -102,1 +100,0 @@
-if (process.env.MCP_REQUIRE_FORWARD_TOKEN === "1" && !forwardToken()) {
@@ -614,1 +611,0 @@
-forward_token_configured: forwardToken() !== "",
diff --git a/dist/runtime.js b/dist/runtime.js
--- a/dist/runtime.js
+++ b/dist/runtime.js
@@ -155,1 +155,0 @@
-/** The header that carries the forward token, and the one that names the client (core-backend waf/mcpforward.go). */
@@ -159,1 +158,0 @@
-export function forwardToken() {
@@ -187,1 +185,0 @@
-const token = forwardToken();

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 3.2.2. If the behavior is intended, declare it (host under permissions.network, a line in the README) instead of removing it; the next crawl scores declarations.

Issue / PR body

Title: echelongraph-mcp: Credential pass-through configuration (KillerSkills verdict caution 71)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.8.1) and scored it **CAUTION 71/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/npm/echelongraph-mcp/3.2.2

## Findings with a location
- `dist/index.js:122` · Credential pass-through configuration — dist/index.js:122 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/http.js:614` · Credential pass-through configuration — dist/http.js:614 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/http.js:32` · Credential pass-through configuration — dist/http.js:32 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/runtime.js:155` · Credential pass-through configuration — dist/runtime.js:155 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/index.js:225` · Credential pass-through configuration — dist/index.js:225 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/runtime.js:187` · Credential pass-through configuration — dist/runtime.js:187 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/http.js:102` · Credential pass-through configuration — dist/http.js:102 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/http.js:39` · Credential pass-through configuration — dist/http.js:39 · Credential pass-through configuration — agent relays sensitive credentials
- `dist/runtime.js:159` · Credential pass-through configuration — dist/runtime.js:159 · Credential pass-through configuration — agent relays sensitive credentials

## Proposed change
The attached patch removes the 9 lines above. If the behavior is intended, the fix is to declare it instead: list the host under `permissions.network` (or the MCP manifest) and say in the README why the package needs it; the next crawl will re-score it with the declaration taken into account.

## How this will be verified
Nothing is "verified safe" by assertion. When a new version is published, KillerSkills re-runs it in the honeypot and the verdict page updates; you can trigger it early from the verdict page ("Re-scan after I publish a fix"). Reply here or at [email protected] if you believe the verdict is wrong; disputes are reviewed by a person and the outcome is published on the verdict page.

_Proposed through killerskills.ai/threats by a user of the package, under their own GitHub account. The evidence is public; the package was never modified by KillerSkills._