Propose a patch

Evidence

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

diff --git a/src/memoryguard/adapters.py b/src/memoryguard/adapters.py
--- a/src/memoryguard/adapters.py
+++ b/src/memoryguard/adapters.py
@@ -571,1 +571,0 @@
-errors.append("memory.md missing after install")
diff --git a/docs/install-codex.md b/docs/install-codex.md
--- a/docs/install-codex.md
+++ b/docs/install-codex.md
@@ -48,1 +48,0 @@
-## Step 3: Add the AGENTS.md instruction
@@ -51,1 +50,0 @@
-its native memory. Add this to your `AGENTS.md` - either the project-level
diff --git a/docs/install-claude-code.md b/docs/install-claude-code.md
--- a/docs/install-claude-code.md
+++ b/docs/install-claude-code.md
@@ -54,1 +54,0 @@
-## Step 3: Add the CLAUDE.md instruction
diff --git a/src/memoryguard/static/cytoscape.min.js b/src/memoryguard/static/cytoscape.min.js
--- a/src/memoryguard/static/cytoscape.min.js
+++ b/src/memoryguard/static/cytoscape.min.js
@@ -32,1 +32,0 @@
-…v=u[g],y=this.calculateLabelDimensions(e,v).width;if(d){var m=v.split("").join("​");v=m}if(y>c){for(var b=v.split(p),x="",w=0;w<b.length;w++){var E=b[w],k=0===x.…

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.2. 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: memoryguard: Instructs the agent to modify its persistent memory or instruction files (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/clawhub/memoryguard/1.0.2

## Findings with a location
- `src/memoryguard/adapters.py:571` · Instructs the agent to modify its persistent memory or instruction files — src/memoryguard/adapters.py:571 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `docs/install-codex.md:51` · Instructs the agent to modify its persistent memory or instruction files — docs/install-codex.md:51 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `docs/install-claude-code.md:54` · Instructs the agent to modify its persistent memory or instruction files — docs/install-claude-code.md:54 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `docs/install-codex.md:48` · Instructs the agent to modify its persistent memory or instruction files — docs/install-codex.md:48 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `src/memoryguard/static/cytoscape.min.js:32` · Hidden zero-width characters detected — src/memoryguard/static/cytoscape.min.js:32 · Hidden zero-width characters detected — may conceal content

## 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._