Propose a patch

Evidence

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

diff --git a/mcp_factory/cli/__init__.py b/mcp_factory/cli/__init__.py
--- a/mcp_factory/cli/__init__.py
+++ b/mcp_factory/cli/__init__.py
@@ -1,1 +1,0 @@
-"""MCP Factory CLI — Main entry point."""
diff --git a/mcp_factory/generator/api_registry.py b/mcp_factory/generator/api_registry.py
--- a/mcp_factory/generator/api_registry.py
+++ b/mcp_factory/generator/api_registry.py
@@ -611,1 +611,0 @@
-# Open .env and paste your API key
diff --git a/mcp_factory/generator/engine.py b/mcp_factory/generator/engine.py
--- a/mcp_factory/generator/engine.py
+++ b/mcp_factory/generator/engine.py
@@ -1,1 +1,0 @@
-"""MCP Generator Engine — Analyzes prompts and generates production-quality MCP server code.
diff --git a/mcp_factory/cli/main.py b/mcp_factory/cli/main.py
--- a/mcp_factory/cli/main.py
+++ b/mcp_factory/cli/main.py
@@ -1,1 +1,0 @@
-"""MCP Factory CLI — Main entry point (re-export)."""
diff --git a/mcp_factory/llm/reviewer.py b/mcp_factory/llm/reviewer.py
--- a/mcp_factory/llm/reviewer.py
+++ b/mcp_factory/llm/reviewer.py
@@ -1,1 +1,0 @@
-"""Code Reviewer — LLM-powered review of generated MCP server code.
diff --git a/tests/test_generator.py b/tests/test_generator.py
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -1,1 +1,0 @@
-"""Tests for MCP Generator Engine."""
diff --git a/mcp_factory/llm/__init__.py b/mcp_factory/llm/__init__.py
--- a/mcp_factory/llm/__init__.py
+++ b/mcp_factory/llm/__init__.py
@@ -1,1 +1,0 @@
-"""LLM module — Provides structured LLM calls for MCP Factory."""
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -309,1 +309,0 @@
-mcpfactory create <prompt>     Generate a new MCP server
diff --git a/mcp_factory/generator/api_tools.py b/mcp_factory/generator/api_tools.py
--- a/mcp_factory/generator/api_tools.py
+++ b/mcp_factory/generator/api_tools.py
@@ -1,1 +1,0 @@
-"""API-specific tool templates — Real working tools for known APIs.
diff --git a/mcp_factory/llm/interactive.py b/mcp_factory/llm/interactive.py
--- a/mcp_factory/llm/interactive.py
+++ b/mcp_factory/llm/interactive.py
@@ -1,1 +1,0 @@
-"""Interactive Prompt Refiner — Asks follow-up questions to improve vague prompts.

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.5.1. 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: prompt2mcp: Hidden zero-width characters detected (KillerSkills verdict caution 68)

## 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 68/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/pypi/prompt2mcp/0.5.1

## Findings with a location
- `mcp_factory/cli/__init__.py:1` · Hidden zero-width characters detected — mcp_factory/cli/__init__.py:1 · Hidden zero-width characters detected — may conceal content
- `mcp_factory/generator/api_registry.py:611` · Prompting user to enter sensitive credentials — mcp_factory/generator/api_registry.py:611 · Prompting user to enter sensitive credentials — social engineering
- `mcp_factory/generator/engine.py:1` · Hidden zero-width characters detected — mcp_factory/generator/engine.py:1 · Hidden zero-width characters detected — may conceal content
- `mcp_factory/cli/main.py:1` · Hidden zero-width characters detected — mcp_factory/cli/main.py:1 · Hidden zero-width characters detected — may conceal content
- `mcp_factory/llm/reviewer.py:1` · Hidden zero-width characters detected — mcp_factory/llm/reviewer.py:1 · Hidden zero-width characters detected — may conceal content
- `tests/test_generator.py:1` · Hidden zero-width characters detected — tests/test_generator.py:1 · Hidden zero-width characters detected — may conceal content
- `mcp_factory/llm/__init__.py:1` · Hidden zero-width characters detected — mcp_factory/llm/__init__.py:1 · Hidden zero-width characters detected — may conceal content
- `README.md:309` · Fake instruction boundary tag — README.md:309 · Fake instruction boundary tag · informational: not an instruction file
- `mcp_factory/generator/api_tools.py:1` · Hidden zero-width characters detected — mcp_factory/generator/api_tools.py:1 · Hidden zero-width characters detected — may conceal content
- `mcp_factory/llm/interactive.py:1` · Hidden zero-width characters detected — mcp_factory/llm/interactive.py:1 · Hidden zero-width characters detected — may conceal content

## Proposed change
The attached patch removes the 10 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._