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. 75 · Caution

Evidence

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

Diff · removes 6 caught lines

Download .patch

diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -110,1 +110,0 @@
-<summary><b>🤖 Using with Kilo Code / Kilo CLI</b></summary> 1. **Auto Model Detection**: Swarm Engine reads your `~/.config/kilo/kilo.jsonc` file automatically
@@ -119,1 +118,0 @@
-<summary><b>🟣 Using with Claude Code</b></summary> 1. The skill is installed at `~/.claude/skills/swarm-engine`. 2. Claude Code will automatically invoke `swar
@@ -127,1 +125,0 @@
-<summary><b>💻 Using with Cursor</b></summary> 1. Open your integrated terminal in Cursor (`Ctrl + ~`). 2. Run `swarm-engine run "your task"` in the terminal. 3
@@ -200,1 +197,0 @@
-<summary><b>Q: Will Swarm Engine mess up my Git branches or commit history?</b></summary> **No.** Swarm Engine was specifically engineered to protect your Git r
@@ -217,1 +213,0 @@
-<summary><b>Q: How does it know which model to use?</b></summary> Swarm Engine contains an intelligent model detector ([`src/detector.ts`](src/detector.ts)): - 
@@ -226,1 +221,0 @@
-<summary><b>Q: Do I need to configure MCP (Model Context Protocol)?</b></summary> **No!** While Swarm Engine includes a full MCP server (`swarm-engine mcp`), we

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: swarm-engine: Collapsed <details> block carrying instructions (KillerSkills verdict caution 75)

## 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 75/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/glama/swarm-engine/1.0.0

## Findings with a location
- `README.md:110` · Collapsed <details> block carrying instructions — README.md:110 · Collapsed <details> block carrying instructions
- `README.md:119` · Collapsed <details> block carrying instructions — README.md:119 · Collapsed <details> block carrying instructions
- `README.md:217` · Collapsed <details> block carrying instructions — README.md:217 · Collapsed <details> block carrying instructions
- `README.md:200` · Collapsed <details> block carrying instructions — README.md:200 · Collapsed <details> block carrying instructions
- `README.md:127` · Collapsed <details> block carrying instructions — README.md:127 · Collapsed <details> block carrying instructions
- `README.md:226` · Collapsed <details> block carrying instructions — README.md:226 · Collapsed <details> block carrying instructions

## 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._

Open the issue on GitHub (PRIYAGDEV/swarm-engine)Fork to open a PR

# or from a terminal, under your own account:
gh repo fork PRIYAGDEV/swarm-engine --clone && cd swarm-engine
curl -s https://killerskills.ai/skills/glama/swarm-engine/1.0.0/patch.diff | git apply --unidiff-zero
git checkout -b killerskills/1.0.0 && git commit -am "swarm-engine: Collapsed <details> block carrying instructions"
gh pr create --fill --body-file <(curl -s https://killerskills.ai/skills/glama/swarm-engine/1.0.0/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.