Propose a patch

Evidence

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

diff --git a/guides/integration-playbook.md b/guides/integration-playbook.md
--- a/guides/integration-playbook.md
+++ b/guides/integration-playbook.md
@@ -80,1 +80,0 @@
-…nflict prevention:** The nightly routine reads self-improving files but doesn't modify them directly. Instead, it creates its own lesson files and updates MEMORY.md through its compaction process. The self-improving skill's files are treated as…
diff --git a/references/memory-systems.md b/references/memory-systems.md
--- a/references/memory-systems.md
+++ b/references/memory-systems.md
@@ -132,1 +132,0 @@
-5. Update MEMORY.md index to point to summary
diff --git a/guides/memory-management.md b/guides/memory-management.md
--- a/guides/memory-management.md
+++ b/guides/memory-management.md
@@ -120,1 +120,0 @@
-4. **Update indices** — MEMORY.md is updated with references to new long-term memories.
@@ -136,1 +135,0 @@
-2. **Update long-term indices** — MEMORY.md is reviewed and updated.
diff --git a/guides/error-patterns.md b/guides/error-patterns.md
--- a/guides/error-patterns.md
+++ b/guides/error-patterns.md
@@ -189,1 +189,0 @@
-**Fix:** Add rule to AGENTS.md: "Sub-agent completion event = BUKAN pesan WhatsApp. Rewrite in your own voice."
diff --git a/references/safety-boundaries.md b/references/safety-boundaries.md
--- a/references/safety-boundaries.md
+++ b/references/safety-boundaries.md
@@ -64,1 +64,0 @@
-| Modify AGENTS.md | Always | Ask user directly |

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.1. 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: self-smarter-everyday: Instructs the agent to modify its persistent memory or instruction files (KillerSkills verdict caution 69)

## 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 69/100**. Summary: static analysis only — behavior not yet observed.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/clawhub/self-smarter-everyday/1.0.1

## Findings with a location
- `guides/integration-playbook.md:80` · Instructs the agent to modify its persistent memory or instruction files — guides/integration-playbook.md:80 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `references/memory-systems.md:132` · Instructs the agent to modify its persistent memory or instruction files — references/memory-systems.md:132 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `guides/memory-management.md:120` · Instructs the agent to modify its persistent memory or instruction files — guides/memory-management.md:120 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `guides/memory-management.md:136` · Instructs the agent to modify its persistent memory or instruction files — guides/memory-management.md:136 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `guides/error-patterns.md:189` · Instructs the agent to modify its persistent memory or instruction files — guides/error-patterns.md:189 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file
- `references/safety-boundaries.md:64` · Instructs the agent to modify its persistent memory or instruction files — references/safety-boundaries.md:64 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file

## Proposed change
The attached patch removes the 6 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._