Security audit
Security checks for vulnerabilities and agentic risk
The bundled skill checks writing locally and optionally saves a style profile, with no evidence of hidden access or data uploads.
Prefer the bundled analyzer; pin any separately downloaded installer or CLI to a reviewed version. Choose writing samples deliberately and remove the saved profile if no longer wanted. Material scanner gaps were investigated manually; separately fetched releases and the hosted demo are outside this artifact review.
Referenced artifact was not completely inspected
- Match a user's personal writing style using a local voice fingerprint profile (`scripts/fingerprint.mjs`).
Referenced artifact was not completely inspected
1. Resolve `<skill_dir>` from this `SKILL.md`; never assume the current working directory.
The skill uses 'network' capability that is not covered by the declared 'allowed-tools' preapproval. Review the host runtime's actual enforcement policy; runtime permissions are unknown unless separately verified, and this declaration alone does not establish a permission ceiling.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
Although this references a CLI rather than an MCP server, the advertised unversioned npx command can fetch and execute a changing npm release when no suitable local package exists. A compromised future release could run with the agent's permissions; claims of prior audits and local-only processing do not establish the safety of that release.
The installation command executes an unpinned skills package and also selects a skill repository without an immutable revision. Although this is not an MCP server, changes or compromise of either upstream could introduce code or agent instructions that were not covered by the stated audits.
This runnable example can download and execute an unpinned humanizer-lint release, despite the scanner incorrectly labeling it an MCP server. A compromised release would receive the draft file and could access other resources available to the executing user, so the local-processing claim does not eliminate the supply-chain risk.
Piping the draft to an unversioned npx command can expose its contents to a newly downloaded, unreviewed package release. This is a CLI supply-chain issue rather than an MCP configuration issue, and a malicious release would execute with the caller's permissions.
The summary advertises an unpinned npx CLI, which can fetch a changing npm release rather than the reviewed version of the tool. The MCP label is inaccurate, but the supply-chain exposure remains; audit and zero-network claims do not constrain the behavior of future releases.
The installation example leaves both the npm installer version and the skill repository revision unspecified. A compromised installer could execute code, while a changed skill revision could supply unreviewed agent instructions; neither risk is resolved by the document's audit claims.
The file-analysis command can fetch and run an unreviewed humanizer-lint release because it specifies no version. Although it is not an MCP server, a compromised CLI release could read the supplied draft and other accessible files or transmit data despite the advertised local-only behavior.
This example supplies draft contents directly to an unpinned npm executable that npx may download at runtime. The actual concern is CLI supply-chain integrity, not MCP: a compromised release could capture the input and execute additional operations with the user's permissions.