Propose a patch

Evidence

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

diff --git a/scripts/UPGRADE_SOP.md b/scripts/UPGRADE_SOP.md
--- a/scripts/UPGRADE_SOP.md
+++ b/scripts/UPGRADE_SOP.md
@@ -30,1 +30,0 @@
-If using an OpenClaw agent, add the sync rule to AGENTS.md:
diff --git a/templates/MODEL_GROUND_TRUTH.md b/templates/MODEL_GROUND_TRUTH.md
--- a/templates/MODEL_GROUND_TRUTH.md
+++ b/templates/MODEL_GROUND_TRUTH.md
@@ -4,1 +4,0 @@
-Usage: After each upgrade, run /verify against this file.
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -92,1 +92,0 @@
-### 2. Add the sync rule to AGENTS.md
@@ -94,1 +93,0 @@
-Add this block to your `AGENTS.md` so the agent reminds you to update ground truth when config changes:
diff --git a/SKILL.md b/SKILL.md
--- a/SKILL.md
+++ b/SKILL.md
@@ -58,1 +58,0 @@
-4. Add the GROUND_TRUTH sync rule to your AGENTS.md (see README)

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.3.6. 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: ground-control: Instructs the agent to modify its persistent memory or instruction files (KillerSkills verdict caution 63)

## 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 63/100**. Summary: static analysis only — behavior not yet observed.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/clawhub/ground-control/0.3.6

## Findings with a location
- `scripts/UPGRADE_SOP.md:30` · Instructs the agent to modify its persistent memory or instruction files — scripts/UPGRADE_SOP.md:30 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `templates/MODEL_GROUND_TRUTH.md:4` · HTML comment carrying imperative instructions to the agent — templates/MODEL_GROUND_TRUTH.md:4 · HTML comment carrying imperative instructions to the agent
- `README.md:92` · Instructs the agent to modify its persistent memory or instruction files — README.md:92 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `README.md:94` · Instructs the agent to modify its persistent memory or instruction files — README.md:94 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `SKILL.md:58` · Instructs the agent to modify its persistent memory or instruction files — SKILL.md:58 · Instructs the agent to modify its persistent memory or instruction files

## 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._