devtools@0.53.0
CAUTION· startup onlyPublisher claims: "MCP Server and CLI for IronBee DevTools" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Startup only; authenticated tool behavior not tested.- Tested
- install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
- Not tested
- the tool list (the server exited before answering) · what the tools do when called with real credentials
What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
POSTHOG_API_KEYConfiguration, not credentials: ALL_KEYWORDS, BROWSER_DEVTOOLS_INSTALL_CHROMIUM, BROWSER_DEVTOOLS_INSTALL_CHROMIUM_HEADLESS, BROWSER_DEVTOOLS_INSTALL_FFMPEG, BROWSER_DEVTOOLS_INSTALL_FIREFOX, BROWSER_DEVTOOLS_INSTALL_WEBKIT, CLAUDE_DESKTOP, CURSOR_TRACE_ID, DEBUG_ENABLE, HOMEDRIVE, PLATFORM, PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD, TELEMETRY_ENABLE- Where it talks to
- no vendor host named in the codeAlso figma.com, linear.app, posthog.com, client-hints.invalid — not the publisher's domain; check what for before connecting.
- What the honeypot saw
- no credential files read · no network
- What we could not test
- the tool list (the server exited before answering) · what the tools do when called with real credentials
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names figma.com, linear.app, posthog.com, client-hints.invalid as well as its own vendor. The guard injects POSTHOG_API_KEY only into requests to the host you bind (and its subdomains) and refuses the rest.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-08 and scored it 74/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Privileged sudo command; Hidden zero-width characters detected; Collapsed <details> block carrying instructions. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 66 s · 3 scenarios · 0 deviationsStatic findings
…717 Update failed.\x1B[0m`),_error("Try running manually with sudo:"),_error(` sudo npm install -g ${packageName}@latest`),process.exit(1)}}catch(err){opts.json?pri…Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.…not read ${resolved} (${code})`)}let parsed;try{parsed=JSON.parse(raw.replace(/^ /,""))}catch{throw new Error(`--secrets-file: ${resolved} is not valid JSON`)}le…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.<summary><strong>Claude Code</strong></summary> Run the following command. See [Claude Code MCP docs](https://docs.anthropic.com/en/docs/claude-code/mcp) for moFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><strong>Smithery</strong></summary> Run the following command. You can find your Smithery API key [here](https://smithery.ai/account/api-keys). See theFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.…_output(` timeout = ${opts.timeout??DEFAULT_TIMEOUT}`),_output(` json = ${opts.json??!1}`),_output(` quiet = ${opts.quiet??!1}`),_output(` verbose = ${opts.verbose??!1}`),_output(` Tip: Start interactive mode with options:`),_output(` ${cl…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…[0]?.toLowerCase()??"";return COMPOSE_PLATFORM_BY_PREFIX[firstSegment]}function mcpServerForPlatform(platformId){return`${platformId}-devtools`}var HTTP_SESSION_SHARE_ENABLE=_envBool("HTTP_SESSION_SHARE_ENABLE")??!1,SESSION_IDLE_SECONDS=_e…Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.…e);tool||(tool=tools.find(t=>t.name().split("_")[1]===resolvedName)),tool||(opts.json?printOutput({status:"not_found",toolName:resolvedName},!0,!0):_error(`Tool '${resolvedName}' not found`),process.exit(1));let inputSchema=tool.inputSchem…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…accessToken:FIGMA_ACCESS_TOKEN?"***":void 0,apiBaseUrl:FIGMA_API_BASE_URL}};opts.json?printOutput(configValues,!0):(_output(`Current Configuration: `),_output(" Daemon:"),_output(` Host: …Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why."inspector": "npm run build && chmod 755 './dist/index.js' && npx @modelcontextprotocol/inspector dist/index.js",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.…timeout(opts.timeout??DEFAULT_TIMEOUT)});if(response.ok){let info=await response.json();_output(`Session: ${info.id}`),_output(`Created: ${new Date(info.createdAt).toISOString()}`),_output(`Last Active: ${new Date(info.lastActiveAt).toISOS…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…ription("Verify the package and install missing Playwright browsers, then exit (npx cache warm-up).").option("--browsers <csv>","comma-separated browsers to ensure: chro…Fix: Pin versions and sources; vendor anything fetched at install time.…nit,mountInternalSurface,parseSeedPayload,prewarmBrowserIfEnabled,shutdown,trackMCPServerStarted,trackToolCalled}from"./core-L3RPPD5C.js";import{Execute,Flush,ToolRegistry,augmentToolInputSchema,createScenarioRunTool,isToolEnabled,jiraTool…Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.…tent-Type","Authorization","MCP-Protocol-Version"]})),app.get("/health",ctx=>ctx.json({status:"ok"})),app.get("/ping",ctx=>ctx.json({status:"ok",message:"pong"})),app.get("/mcp",ctx=>ctx.json({status:"ok",protocol:"model-context-protocol",…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…amableHTTPTransport}from"@hono/mcp";import{serve}from"@hono/node-server";import{McpServer as MCPServer}from"@modelcontextprotocol/sdk/server/mcp.js";import{StdioServerTransport}from"@modelcontextprotocol/sdk/server/stdio.js";import{Hono}fr…Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.…n?JSON.stringify(data,null,2):String(data);isError?console.error(output):console.log(output)}function getZodTypeName(schema){let typeName=schema._def.typeName;return typeName==="ZodOptional"||typeName==="ZodNullable"||typeName==="ZodDefaul…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…sync ctx=>{let reqClone=ctx.req.raw.clone();debug(`Got request: ${await reqClone.json()}`)};app.use("*",daemonRequestGuard({host:daemonHost})),app.get("/health",ctx=>ctx.json({status:"ok"})),app.get("/info",ctx=>{let info2={version:require…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…rded(token);app.post("/internal/secrets",async ctx=>{if(!guarded(ctx))return ctx.json(NOT_FOUND,404);let body;try{body=await ctx.req.json()}catch{return error("Secret seed refused: body is not valid JSON"),ctx.json({error:"Invalid secrets:…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.…n:"),_output(` npm install -g ${packageName}@latest`),_output("or"),_output(` npx ${packageName}@latest`);return}let rl=readline.createInterface({input:process.stdin,output:process.stdout}),answer…Fix: Pin versions and sources; vendor anything fetched at install time.Cost to the agent
Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.
Known vulnerabilities
No advisories in OSV for @ironbee-ai/devtools@0.53.0 as of 6d ago. This is the package itself, not its dependencies.
What would make this Verified
6 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim ironbee-ai: 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.
- Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
- File type combined with upload action≈ +2 on the totalRemove the call, or declare the host in frontmatter and explain in the README what is sent and why.
- MCP server using network transport (SSE/HTTP)≈ +2 on the totalNamespace tool names and keep tool descriptions literal; no instructions to the agent inside them.