kxm@0.7.216
CAUTION· install onlyPublisher claims: "KXM local-first multi-agent orchestration and operator dashboard" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Nothing executable in the bundle ran; the trace covers the install step only.- Tested
- every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-13 and scored it 65/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; Collapsed <details> block carrying instructions; Privileged sudo command. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 1 scenarios · 0 deviationsStatic findings
<summary>PowerShell</summary> ```powershell # Step 1: ignore runtime state Add-Content .gitignore ".kxm/state/", ".kxm/logs/", ".kxm/backups/" # Step 2: start tFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>PowerShell</summary> ```powershell $env:KXM_SERVER_URL = "http://127.0.0.1:7331" $env:KXM_AUTH_TOKEN = "replace-with-the-project-token" $env:KXM_WORKDIFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.const chmod = deps.chmodSync ?? chmodSync;Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.| Platform-wide | `svc-<system>-<purpose>` |Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.* and update the AGENTS.md command block. Owned skills are replaced; unrelatedFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.sudo -n DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends "${packages[@]}"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell."test:packages": "npx nx run-many -t test --output-style=static",Fix: Pin versions and sources; vendor anything fetched at install time..option("--sudo", "Use sudo privileges on remote file")Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.<summary>PowerShell</summary> ```powershell $env:KXMD_HUB_TOKEN = op read "op://Private/kxmd/admin-token" kxm hub bind --cloud --token-env KXMD_HUB_TOKEN http:/Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct."build:packages": "npx nx run-many -t build --output-style=stream"Fix: Pin versions and sources; vendor anything fetched at install time.<summary>PowerShell</summary> ```powershell Set-Location C:\srv\kxm\product $env:KXM_HOST = "127.0.0.1" $env:KXM_PORT = "7331" $env:KXM_AUTH_TOKEN = Get-ContentFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.| Tenant-scoped | `svc-<tenant>-<system>-<purpose>` |Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.print(runtime.io, runtime.json, payload, JSON.stringify(output, null, 2));Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why..option("--sudo", "Execute remote command with sudo privileges")Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.chmod 600 "$pwfile"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.sudo -n apt-get updateFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.const chmod = options.chmodSync ?? chmodSync;Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts."check:packages": "npx nx run-many -t build,test:coverage,typecheck --output-style=static",Fix: Pin versions and sources; vendor anything fetched at install time.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
No advisories in OSV for @kontextmind/kxm@0.7.216 as of 30 h ago. This is the package itself, not its dependencies.
What would make this Verified
15 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Text that tells the agent to ignore its instructions≈ +9 on the totalDelete phrases like "ignore previous instructions", "override", "bypass" from SKILL.md and tool descriptions. Describe what the skill does; do not instruct the agent about other rules.
- Unclaimed publisher≈ +8 on the totalClaim kontextmind: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
- Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.