Propose a patch
A fix the maintainer can act on, with the evidence attached. It opens under your GitHub account in your browser; KillerSkills holds no token and never modifies the package. The fix is verified the only way we verify anything: the next crawl re-runs the new version in the honeypot. 78 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 7 caught lines
Download .patchdiff --git a/assets/AGENTS.md b/assets/AGENTS.md --- a/assets/AGENTS.md +++ b/assets/AGENTS.md @@ -81,1 +81,0 @@ -**Proactive work (do without asking):** Read/organise memory, check projects, update docs, review and maintain MEMORY.md. diff --git a/references/compaction-survival.md b/references/compaction-survival.md --- a/references/compaction-survival.md +++ b/references/compaction-survival.md @@ -32,1 +32,0 @@ -Add this to your AGENTS.md: @@ -94,1 +93,0 @@ -To support this, add to AGENTS.md: diff --git a/assets/anchor.md b/assets/anchor.md --- a/assets/anchor.md +++ b/assets/anchor.md @@ -3,1 +3,0 @@ -This file exists because context compaction can silently remove instructions from your agent's working memory mid-session. Keep this under 500 tokens. Your agen diff --git a/references/setup-guide.md b/references/setup-guide.md --- a/references/setup-guide.md +++ b/references/setup-guide.md @@ -83,1 +83,0 @@ -…e Your agent should periodically (every few days) review recent daily files and update MEMORY.md. Add this to AGENTS.md if not already there. Old daily files can be archived or deleted after their in… diff --git a/references/memory-architecture.md b/references/memory-architecture.md --- a/references/memory-architecture.md +++ b/references/memory-architecture.md @@ -129,1 +129,0 @@ -3. **Update MEMORY.md** with distilled insights @@ -138,1 +137,0 @@ -- Update MEMORY.md with new insights
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 1.0.3. 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: kickstart: Instructs the agent to modify its persistent memory or instruction files (KillerSkills verdict caution 78)
## What the honeypot observed KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.9.2) and scored it **CAUTION 78/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/clawhub/kickstart/1.0.3 ## Findings with a location - `assets/AGENTS.md:81` · Instructs the agent to modify its persistent memory or instruction files — assets/AGENTS.md:81 · Instructs the agent to modify its persistent memory or instruction files - `references/compaction-survival.md:32` · Instructs the agent to modify its persistent memory or instruction files — references/compaction-survival.md:32 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `assets/anchor.md:3` · HTML comment carrying imperative instructions to the agent — assets/anchor.md:3 · HTML comment carrying imperative instructions to the agent - `references/compaction-survival.md:94` · Instructs the agent to modify its persistent memory or instruction files — references/compaction-survival.md:94 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `references/setup-guide.md:83` · Instructs the agent to modify its persistent memory or instruction files — references/setup-guide.md:83 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `references/memory-architecture.md:138` · Instructs the agent to modify its persistent memory or instruction files — references/memory-architecture.md:138 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `references/memory-architecture.md:129` · Instructs the agent to modify its persistent memory or instruction files — references/memory-architecture.md:129 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file ## Proposed change The attached patch removes the 7 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._
We count proposals opened, not merged: a merge shows up as a new version that the crawl re-runs. Disputes: the maintainer can contest the verdict from its page; a person reviews it and the outcome is published.