Security audit
Security checks for vulnerabilities and agentic risk
This skill provides a clearly limited workflow for organizing supplied product records into reviewable drafts without accessing or changing a shop backend.
Share only product records you intend your agent to process and review the CSV before importing it. If using the documented npx installation command, verify the source and prefer reviewed installer versions and source revisions.
The only supplied executable behavior is an external skill installation command, which is materially different from the declared data-normalization workflow. The remote skill's implementation is not included, so its behavior and compliance with the declared processing constraints cannot be verified. Nothing shown establishes shop-backend access or other malicious behavior.
The documented installation command uses an unpinned npm CLI (npx skills), allowing its resolved version to change; if that package or a future release is compromised, running the command could execute attacker-controlled code with the installer's permissions. This is a supply-chain exposure rather than an MCP-server configuration issue, and its placement in optional installation documentation makes it less immediate than automatic execution during SKU processing. Nothing in the supplied file establishes malicious intent or an actual compromise.
The installation command invokes npx skills without specifying a version, potentially fetching and executing a changing npm package with the user's permissions. A compromised package release could therefore execute malicious code during installation, although the README contains no evidence of such a compromise. The finding incorrectly describes this as an MCP server reference; this is an installation-time supply-chain risk, not a requirement of the SKU-normalization workflow.
Without declared permissions the skill's intent is opaque and cannot be validated.
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
permit persons to whom the Software is furnished to do so.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
No suspicious patterns detected.