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. 69 · Caution
Evidence
The sandbox observed nothing beyond the static findings; the proposal cites those.
Diff · removes 7 caught lines
Download .patchdiff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -29,1 +29,0 @@ -# Boot check (add to AGENTS.md startup) @@ -75,1 +74,0 @@ -Add to your `AGENTS.md` startup sequence: diff --git a/COMPLETION_REPORT.md b/COMPLETION_REPORT.md --- a/COMPLETION_REPORT.md +++ b/COMPLETION_REPORT.md @@ -125,1 +125,0 @@ -### Step 1: Add to AGENTS.md Startup @@ -127,1 +126,0 @@ -Add this line to the "Every Session" section of AGENTS.md: diff --git a/SKILL.md b/SKILL.md --- a/SKILL.md +++ b/SKILL.md @@ -26,1 +26,0 @@ -Add this line to your `AGENTS.md` startup section: @@ -326,1 +325,0 @@ -Add to your `AGENTS.md` startup sequence (after reading SOUL.md, USER.md): diff --git a/TEST_RESULTS.md b/TEST_RESULTS.md --- a/TEST_RESULTS.md +++ b/TEST_RESULTS.md @@ -181,1 +181,0 @@ -1. Add `bash /path/to/editorial/boot-check.sh` to AGENTS.md
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: chitin-chronicle: 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 1.9.2) 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/chitin-chronicle/1.0.0 ## Findings with a location - `README.md:29` · Instructs the agent to modify its persistent memory or instruction files — README.md:29 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `COMPLETION_REPORT.md:127` · Instructs the agent to modify its persistent memory or instruction files — COMPLETION_REPORT.md:127 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `README.md:75` · Instructs the agent to modify its persistent memory or instruction files — README.md:75 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `SKILL.md:326` · Instructs the agent to modify its persistent memory or instruction files — SKILL.md:326 · Instructs the agent to modify its persistent memory or instruction files - `COMPLETION_REPORT.md:125` · Instructs the agent to modify its persistent memory or instruction files — COMPLETION_REPORT.md:125 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file - `SKILL.md:26` · Instructs the agent to modify its persistent memory or instruction files — SKILL.md:26 · Instructs the agent to modify its persistent memory or instruction files - `TEST_RESULTS.md:181` · Instructs the agent to modify its persistent memory or instruction files — TEST_RESULTS.md:181 · 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.