bla-agents@3.3.2
CAUTION· startup onlyPublisher claims: "BLA (Blind LLM Agents) — Adversarial defense shield that protects binaries, Electron apps, JS bundles, Python, WebAssembly, SQLite, AI Models, and Android DEX/IL2CPP against AI reverse-engineering agents like REA." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Startup only; authenticated tool behavior not tested.- Tested
- 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
- no credential variables read in the code
- Where it talks to
- bla.tools (its own)Also local.fixture, schemastore.org, cursor.com, idx.dev, windsurf.com, visualstudio.com, opencode.ai — 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 under the guard. Startup only; authenticated tool behavior not tested. The code names no credential variable and talks to bla.tools, its own; the guard pins the network to what the honeypot saw and logs every call.
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 (2 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-15 and scored it 66/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; Access to sensitive file paths; Destructive file system command detected. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 28 s · 2 scenarios · 0 deviationsStatic findings
<summary><strong>Designed-for paths not empirically run on this machine (click to expand)</strong></summary> <br /> These shields exist in-tree and are covered Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.: "No package.json or .env.production found to verify Node launch flags.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.text: `${f.kind} (${f.representation}) in ${f.file} — Run "npx bla-agents defend" to replace with decoy source maps.`,Fix: Pin versions and sources; vendor anything fetched at install time.…scription: "Read-only correlation of a prior harden run: load bla-harden-summary.json by path or output directory, optionally asserting a run_id. Unsigned summaries only — never treat…Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.* with the bankrupt profile. Modelled on REA's `npx rea-agents setup` simplicity.Fix: Pin versions and sources; vendor anything fetched at install time.… CLI update: npm install --global bla-agents@latest\n • Agent MCP refresh: npx bla-agents@latest setup\n`); } return 0; } …Fix: Pin versions and sources; vendor anything fetched at install time.globalUpdateCommand: "npm install --global bla-agents@latest",Fix: Pin the exact version (`[email protected]`, `==1.2.3`) so every install gets the bytes that were reviewed.` ${ANSI.bold}Format / Preset:${ANSI.reset} ${res.detectedFormat} / ${ANSI.magenta}${res.intensity}${ANSI.reset} (seed: ${res.seed})`,Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.…ude leaked environment artifact \`${rel}\` (SOLUTION/writeup notes, populated \`.env\`, SSH/PEM private keys, or \`.pdb\`/\`.dSYM\`/\`.dwp\` debug databases) from t…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.sources.push(".env.production");Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.… allow bankrupt/maximum harden despite leaked SOLUTION/writeup notes, populated .env secrets, SSH/PEM private keys, or unstripped .pdb/.dSYM/.dwp debug databases. D…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.description: "Treat leaked SOLUTION/writeup notes, populated .env secrets, private keys, or .pdb/.dSYM/.dwp debug symbols as non-blocking (lab only)",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.npx bla-shield <command> [options]Fix: Pin versions and sources; vendor anything fetched at install time."build": "node scripts/sync-version.mjs && tsc -p tsconfig.build.json && chmod +x dist/cli.js",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.<summary><strong>Do I need Ghidra, Hopper, IDA, or Wabt installed to protect files?</strong></summary> No. All 19+ format shields are built into `bla-agents` inFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.* Designed so that once an AI Coding Agent attaches BLA via `npx bla-agents@latest setup`,Fix: Pin versions and sources; vendor anything fetched at install time.: join(rootDir, ".env.production");Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.const recommendedCommand = `npx bla-agents protect ${targetPath}`;Fix: Pin versions and sources; vendor anything fetched at install time.npx bla-agents setup -y # register bla_harden with your agentFix: Pin versions and sources; vendor anything fetched at install time.…r()}\n\nBLA v${BLA_IDENTITY.version} update workflow:\n • Global CLI update: npm install --global bla-agents@latest\n • Agent MCP refresh: npx bla-agents@latest setup\n`); } …Fix: Pin the exact version (`[email protected]`, `==1.2.3`) so every install gets the bytes that were reviewed.console.log(JSON.stringify({ ...emitted, outputPath: outPath }, null, 2));Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.refreshSetupCommand: "npx bla-agents@latest setup",Fix: Pin versions and sources; vendor anything fetched at install time.description: "Optional path to bla-harden-summary.json (or its output directory) to correlate this verify with a prior harden run (unsigned).",Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.exposedFlags.push(`.env.production contains inspector/run-as-node directive`);Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.throw new ReleaseConsistencyError("INVALID_PACKAGE_VERSION", `Invalid package.json version '${input.packageVersion}': expected strict semver (X.Y.Z)`);Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.npx bla-agents <command> [options]Fix: Pin versions and sources; vendor anything fetched at install time.* with the bankrupt profile. Modelled on REA's `npx rea-agents setup` simplicity.Fix: Pin versions and sources; vendor anything fetched at install time.* Designed so that once an AI Coding Agent attaches BLA via `npx bla-agents@latest setup`,Fix: Pin versions and sources; vendor anything fetched at install time.description: "Path to bla-harden-summary.json, or the harden output directory that contains it",Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.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
Not checked against OSV yet; the weekly check has not reached this version.
What would make this Verified
14 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim qybaihe: 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.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Unpinned dependencies≈ +3 on the totalPin exact versions (and ideally hashes) in package.json, requirements or the install script so the bundle cannot change under you.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.