Propose a patch

Evidence

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

diff --git a/src/bot/menu.ts b/src/bot/menu.ts
--- a/src/bot/menu.ts
+++ b/src/bot/menu.ts
@@ -39,1 +39,0 @@
-import { fallBack, relayState, relayToken, tokenHash } from "../relay";
@@ -515,1 +514,0 @@
-const token = relayToken();
diff --git a/src/relay.ts b/src/relay.ts
--- a/src/relay.ts
+++ b/src/relay.ts
@@ -66,1 +66,0 @@
-export const relayToken = () => `relay_${randomId(24)}`;
@@ -178,1 +177,0 @@
-if (!relay) return new Response("unknown relay token\n", { status: 401 });
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -38,1 +38,0 @@
-<summary>Or from a terminal</summary> ```bash bun install ``` ```bash bunx wrangler login ``` ```bash bun run deploy ``` ```bash bunx wrangler secret put BOT_TO
diff --git a/test/relay.test.ts b/test/relay.test.ts
--- a/test/relay.test.ts
+++ b/test/relay.test.ts
@@ -41,1 +41,0 @@
-async function relayToken(): Promise<string> {
@@ -82,1 +81,0 @@
-const token = await relayToken();
@@ -98,1 +96,0 @@
-await relayToken();
@@ -105,1 +102,0 @@
-const token = await relayToken();
@@ -127,1 +123,0 @@
-const token = await relayToken();
diff --git a/src/tools/edit.ts b/src/tools/edit.ts
--- a/src/tools/edit.ts
+++ b/src/tools/edit.ts
@@ -236,1 +236,0 @@
-… 😱 🎉 🤩 🙏 👌 🕊 🤡 🥱 😍 🐳 💯 🤣 ⚡ 🏆 💔 🤨 😐 🍓 🍾 💋 🖕 😈 😴 😭 🤓 👻 👨‍💻 👀 🎃 🙈 😇 😨 🤝 ✍ 🤗 🫡 🎅 🎄 ☃ 💅 🤪 🗿 🆒 💘 🙉 🦄 😘 💊 🙊 😎 👾 🤷 😡. …

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.0. 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: telegram-mcp: Credential pass-through configuration (KillerSkills verdict caution 62)

## 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 62/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/glama/telegram-mcp/1.0.0

## Findings with a location
- `src/bot/menu.ts:515` · Credential pass-through configuration — src/bot/menu.ts:515 · Credential pass-through configuration — agent relays sensitive credentials
- `src/relay.ts:66` · Credential pass-through configuration — src/relay.ts:66 · Credential pass-through configuration — agent relays sensitive credentials
- `src/bot/menu.ts:39` · Credential pass-through configuration — src/bot/menu.ts:39 · Credential pass-through configuration — agent relays sensitive credentials
- `src/relay.ts:178` · Credential pass-through configuration — src/relay.ts:178 · Credential pass-through configuration — agent relays sensitive credentials
- `README.md:38` · Collapsed <details> block carrying instructions — README.md:38 · Collapsed <details> block carrying instructions
- `test/relay.test.ts:41` · Credential pass-through configuration — test/relay.test.ts:41 · Credential pass-through configuration — agent relays sensitive credentials
- `test/relay.test.ts:98` · Credential pass-through configuration — test/relay.test.ts:98 · Credential pass-through configuration — agent relays sensitive credentials
- `src/tools/edit.ts:236` · Hidden zero-width characters detected — src/tools/edit.ts:236 · Hidden zero-width characters detected — may conceal content
- `test/relay.test.ts:127` · Credential pass-through configuration — test/relay.test.ts:127 · Credential pass-through configuration — agent relays sensitive credentials
- `test/relay.test.ts:82` · Credential pass-through configuration — test/relay.test.ts:82 · Credential pass-through configuration — agent relays sensitive credentials
- `test/relay.test.ts:105` · Credential pass-through configuration — test/relay.test.ts:105 · Credential pass-through configuration — agent relays sensitive credentials

## Proposed change
The attached patch removes the 11 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._