Propose a patch

Evidence

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

diff --git a/dist/shell/view.js b/dist/shell/view.js
--- a/dist/shell/view.js
+++ b/dist/shell/view.js
@@ -76,1 +76,0 @@
-Projects: [`${k("↑↓")} ${d("move")}`, `${k("↵")} ${d("open")}`, `${k("e")} ${d("edit AGENTS.md")}`],
diff --git a/dist/init/init.js b/dist/init/init.js
--- a/dist/init/init.js
+++ b/dist/init/init.js
@@ -443,1 +443,0 @@
-log(`  Run \`phren init --apply-starter-update\` to refresh global/AGENTS.md and global skills.`);
diff --git a/dist/link/link.js b/dist/link/link.js
--- a/dist/link/link.js
+++ b/dist/link/link.js
@@ -86,1 +86,0 @@
-log(`  Run \`phren init --apply-starter-update\` to refresh global/AGENTS.md and global skills.`);

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.3.73. 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: Instructs the agent to modify its persistent memory or instruction files (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/0.3.73

## Findings with a location
- `dist/shell/view.js:76` · Instructs the agent to modify its persistent memory or instruction files — dist/shell/view.js:76 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `dist/init/init.js:443` · Instructs the agent to modify its persistent memory or instruction files — dist/init/init.js:443 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `dist/link/link.js:86` · Instructs the agent to modify its persistent memory or instruction files — dist/link/link.js:86 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file

## Proposed change
The attached patch removes the 3 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._