Propose a patch

Evidence

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

diff --git a/scripts/evaluate.py b/scripts/evaluate.py
--- a/scripts/evaluate.py
+++ b/scripts/evaluate.py
@@ -49,1 +49,0 @@
-"name": "Update MEMORY.md with new fact",
@@ -50,1 +49,0 @@
-…ight": 0.08, "name": "Update MEMORY.md with new fact", "task": "Add a new entry to MEMORY.md: 'Hoss ran its first meta-harness evolution on [TODAY DATE]. The loop successfu…
@@ -53,1 +51,0 @@
-0: "Did not update MEMORY.md",
diff --git a/references/benchmark-design.md b/references/benchmark-design.md
--- a/references/benchmark-design.md
+++ b/references/benchmark-design.md
@@ -125,1 +125,0 @@
-- Update MEMORY.md with new decisions

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.0. 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: meta-harness-evolver: Instructs the agent to modify its persistent memory or instruction files (KillerSkills verdict caution 70)

## 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 70/100**. Summary: static analysis only — behavior not yet observed.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/clawhub/meta-harness-evolver/1.0.0

## Findings with a location
- `scripts/evaluate.py:49` · Instructs the agent to modify its persistent memory or instruction files — scripts/evaluate.py:49 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `scripts/evaluate.py:50` · Instructs the agent to modify its persistent memory or instruction files — scripts/evaluate.py:50 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `scripts/evaluate.py:53` · Instructs the agent to modify its persistent memory or instruction files — scripts/evaluate.py:53 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `references/benchmark-design.md:125` · Instructs the agent to modify its persistent memory or instruction files — references/benchmark-design.md:125 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file

## Proposed change
The attached patch removes the 4 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._