Is agents safe?
- Node.js filesystem read
- Node.js filesystem write/delete
- Node.js network egress
agents is an AI npm_package analyzed by SkillTotal's deterministic static scanner. The scan found no malicious indicators, though 4 risky constructs are reported for review. It can: filesystem read, filesystem write, mcp tools detected and network egress — capabilities are what the code can do, not a verdict on intent. Risk score 0/100 (low).
agents 0.24.0
Automated static-analysis result. It can contain false positives and false negatives, and is not a claim about the intent of agents's authors. Report a false positive.
Behavioral traits
How this component maps to the CSA agentic threat model. Descriptive — it never affects the risk score.
Findings (4)
The component reads files from disk.
return computer ? await workspace.fs.readFile(path, "utf8") : await workspace.readFile(path);
return await drain(await workspace.fs.readFile(path));
Why it matters: Usually legitimate, but worth confirming it can't be steered into reading sensitive files.
Fix: Confirm which files are read and that paths cannot be influenced by untrusted input to reach sensitive locations.
The component writes or deletes files on disk.
if (computer) await workspace.fs.writeFile(path, content);
if (computer) await workspace.fs.writeFile(path, resource.content);
if (computer) await workspace.fs.writeFile(path, bytes);
Why it matters: Usually legitimate, but worth confirming the paths can't be controlled by untrusted input.
Fix: Confirm which files are written/deleted and that paths cannot be influenced by untrusted input.
The component makes outbound network requests.
* async fetch(req, env) {* - RPC methods only. `.fetch()` is not supported (will throw).
fetch(request: Request): Promise<Response>;
* previous "construct a Request in the parent DO and `stub.fetch()`
* Facet-parent stubs route normal HTTP `.fetch()` calls through the
const response = await retryDurableObjectOperation(() => namespace.get(id, getOptions).fetch(cloneRequestForFetch(request)), {response = await fetch(`${apiBaseUrl}/conversations.open`, {response = await fetch(`${apiBaseUrl}/${method}`, {response = await fetch(`${apiBaseUrl}/bot${options.botToken}/sendMessage`, {return asApiResponse(await (await fetch(`${apiBaseUrl}/bot${options.botToken}/sendMessageDraft`, {const response = await fetch(messagesUrl, {const response = await fetch(getMessagesUrl.toString(), {static fetch(_opts: PartyFetchOptions): Promise<Response>;
static fetch(_opts) {if (opts.basePath) return PartySocket.fetch({return PartySocket.fetch({fetch(input: RequestInfo | URL, init?: RequestInit): Promise<Response>;
const response = await fetch(endpoint, { headers: options?.headers });const wsResponse = await fetch(fetchUrl, { headers: {const response = await browser.fetch(browserSessionEndpoint(void 0, options), { method: "POST" });const response = await browser.fetch(`https://localhost/v1/devtools/browser/${sessionId}/json/list`);const response = await browser.fetch(`https://localhost/v1/devtools/browser/${sessionId}`, { method: "DELETE" });const response = await browser.fetch(browserSessionEndpoint(void 0, {const ws = (await browser.fetch(browserSessionEndpoint(sessionId), { headers: { Upgrade: "websocket" } })).webSocket;const response = await fetch(endpoint, { headers });Why it matters: Usually legitimate, but confirm the destinations are expected and no sensitive data leaves.
Fix: Confirm the destination hosts are expected and that no sensitive data is sent off-host.
An MCP tool surface (manifest or tool definitions) was found.
* navigator.modelContext?.registerTool({* navigator.modelContext?.registerTool({modelContext.registerTool(toolDef, { signal: controller.signal });{"version":3,"file":"webmcp.js","names":[],"sources":["../../src/experimental/webmcp.ts"],"sourcesContent":["/**\n * !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n * !! WARNING: EXPERIMENTAL — DO NOT USE IN PRODU …{"version":3,"file":"webmcp.js","names":[],"sources":["../../src/experimental/webmcp.ts"],"sourcesContent":["/**\n * !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!\n * !! WARNING: EXPERIMENTAL — DO NOT USE IN PRODU …return server.registerTool(name, {{"version":3,"file":"x402.js","names":[],"sources":["../../../src/mcp/client/x402.ts"],"sourcesContent":["/**\n * X402 MCP Integration (v2)\n *\n * Based on:\n * - Coinbase's x402 (Apache 2.0): https://github.com/coinbase/x402\n * - @ethann …Why it matters: Just context — review which tools it offers and their permissions.
Fix: Review the declared MCP tools and their permissions.
Check your own component
Run the same evidence-backed scan on any MCP server, agent skill, or package.
Scan your own componentHow we determine this: deterministic static analysis (regex + AST), evidence-anchored, no code execution. Methodology →