← rozo-checkout@0.1.25
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. 76 · 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 @@ -243,1 +243,0 @@ -<summary><b>Claude Code</b> — install the skill, or paste the one-liner</summary> This repo is a Claude Code skill: it ships `SKILL.md` plus the executables in @@ -265,1 +264,0 @@ -<summary><b>Codex CLI</b> — AGENTS.md, or run it directly</summary> Codex reads `AGENTS.md` from the project root. Add a standing instruction so it knows how to @@ -281,1 +279,0 @@ -<summary><b>OpenCode</b> — AGENTS.md, or run it directly</summary> OpenCode also reads `AGENTS.md` from the project root, so the Codex snippet above works uncha @@ -296,1 +293,0 @@ -<summary><b>Cline</b> — .clinerules, or run it directly</summary> Cline reads standing instructions from `.clinerules` in the project root: ``` To pay an OpenRo @@ -311,1 +307,0 @@ -<summary><b>Cursor</b> — .cursor/rules, or run it directly</summary> Add a project rule at `.cursor/rules/rozo-checkout.mdc`: ``` To pay an OpenRouter / Coinbas @@ -326,1 +321,0 @@ -<summary><b>Hermes Agent</b> — run the one-liner in a session</summary> Hermes Agent (Nous Research) has shell access and its own skill system. Start it with `h @@ -357,1 +351,0 @@ -<summary><b>Pi</b> — run the one-liner in a session</summary> Pi is a BYOK terminal agent whose built-in tools include `bash`, so it can run the command directl
Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.1.25. 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: rozo-checkout: Collapsed <details> block carrying instructions (KillerSkills verdict caution 76)
## 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 76/100**. Summary: no network, touches nothing outside the project, writes no memory. - See the findings below. Full trace and findings: https://killerskills.ai/skills/clawhub/rozo-checkout/0.1.25 ## Findings with a location - `README.md:281` · Collapsed <details> block carrying instructions — README.md:281 · Collapsed <details> block carrying instructions - `README.md:311` · Collapsed <details> block carrying instructions — README.md:311 · Collapsed <details> block carrying instructions - `README.md:357` · Collapsed <details> block carrying instructions — README.md:357 · Collapsed <details> block carrying instructions - `README.md:326` · Collapsed <details> block carrying instructions — README.md:326 · Collapsed <details> block carrying instructions - `README.md:296` · Collapsed <details> block carrying instructions — README.md:296 · Collapsed <details> block carrying instructions - `README.md:265` · Collapsed <details> block carrying instructions — README.md:265 · Collapsed <details> block carrying instructions - `README.md:243` · Collapsed <details> block carrying instructions — README.md:243 · Collapsed <details> block carrying instructions ## 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._
# or from a terminal, under your own account: gh repo fork RozoAI/rozo-checkout-skill --clone && cd rozo-checkout-skill curl -s https://killerskills.ai/skills/clawhub/rozo-checkout/0.1.25/patch.diff | git apply --unidiff-zero git checkout -b killerskills/0.1.25 && git commit -am "rozo-checkout: Collapsed <details> block carrying instructions" gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/clawhub/rozo-checkout/0.1.25/patch.md)
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.