Propose a patch

Evidence

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

diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -298,1 +298,0 @@
-git add .agents/agents.json .agents/skills/ AGENTS.md && git commit -m "Add agents config"
@@ -326,1 +325,0 @@
-<summary><b>Why does Claude Code use <code>.mcp.json</code> now?</b></summary> <br/> Project scope is the location Anthropic documents for teams: it lives in th
@@ -344,1 +342,0 @@
-<summary><b>Can I target an MCP server to specific tools only?</b></summary> <br/> Yes, with <code>"targets"</code> in the definition or <code>--target</code> o
@@ -356,1 +353,0 @@
-<summary><b>How do I upgrade from 0.8.x?</b></summary> <br/> Run any command. <code>.agents/agents.json</code> migrates from schema 3 to 4 automatically and the
diff --git a/dist/core/skills.js b/dist/core/skills.js
--- a/dist/core/skills.js
+++ b/dist/core/skills.js
@@ -221,1 +221,0 @@
-* reaches the bridge. A link to `~/.ssh/id_rsa` would arrive the same way, as a real

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.9.4. 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: cli: Collapsed <details> block carrying instructions (KillerSkills verdict caution 75)

## 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 75/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/npm/cli/0.9.4

## Findings with a location
- `README.md:344` · Collapsed <details> block carrying instructions — README.md:344 · Collapsed <details> block carrying instructions
- `README.md:356` · Collapsed <details> block carrying instructions — README.md:356 · Collapsed <details> block carrying instructions
- `README.md:326` · Collapsed <details> block carrying instructions — README.md:326 · Collapsed <details> block carrying instructions
- `dist/core/skills.js:221` · References a credential file in the home directory — dist/core/skills.js:221 · References a credential file in the home directory
- `README.md:298` · Instructs the agent to modify its persistent memory or instruction files — README.md:298 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file

## Proposed change
The attached patch removes the 5 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._