Skills/npm/tokensafe

tokensafe@4.3.1

CAUTION· startup only

Publisher claims: "Your AI agent is smart enough to code all day. Now it actually can. Local runtime firewall & zero-loss context optimizer for AI coding agents." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher tokensafe-dev · publishing for 13 daysInstalls 20,911Last detonated 1 h ago

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
ANTHROPIC_API_KEYDEEPSEEK_API_KEYGEMINI_API_KEYOPENAI_API_KEYTOKENSAFE_AGENTTOKENSAFE_ALLOWED_ORIGINSTOKENSAFE_AUTOPRUNE_INTERVALTOKENSAFE_BG_LICENSE_KEYTOKENSAFE_BUNDLE_MAX_TOKENSTOKENSAFE_CACHE_TIMEOUTTOKENSAFE_CLOUD_DISTILLTOKENSAFE_CLOUD_URLTOKENSAFE_CONTEXT_WINDOWTOKENSAFE_DATA_DIRTOKENSAFE_DB_PATHTOKENSAFE_GRAPH_DISK_CACHETOKENSAFE_GRAPH_TTLTOKENSAFE_HMAC_SALTTOKENSAFE_HOSTTOKENSAFE_LICENSE_KEYTOKENSAFE_LIVE_FLUSH_SECONDSTOKENSAFE_LIVE_URLTOKENSAFE_MAX_EVENTSTOKENSAFE_MAX_TURN_TOKENSTOKENSAFE_MCP_AUTORELOADTOKENSAFE_MCP_TIMEOUTTOKENSAFE_NONINTERACTIVETOKENSAFE_PRIVATE_KEYTOKENSAFE_PROXY_DISABLEDTOKENSAFE_PUBLIC_KEYTOKENSAFE_RETENTION_DAYSTOKENSAFE_SELF_HOSTEDTOKENSAFE_STALE_GUARDTOKENSAFE_TARGET_TURN_TOKENSTOKENSAFE_TELEMETRY_DISABLEDTOKENSAFE_TEST_PROMPTTOKENSAFE_TOOL_PRUNE_ALLOWLISTTOKENSAFE_TOOL_PRUNE_KEEPTOKENSAFE_TOOL_PRUNINGTOKENSAFE_TREESITTER or --claim-token
Where it talks to
tokensafe.dev (its own)Also deepseek.com, python.org, z.ai, apple.com — not the publisher's domain; check what for before connecting.
What we saw when it ran
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 deepseek.com, python.org, z.ai, apple.com as well as its own vendor. The guard injects ANTHROPIC_API_KEY, DEEPSEEK_API_KEY, GEMINI_API_KEY, OPENAI_API_KEY, TOKENSAFE_AGENT, TOKENSAFE_ALLOWED_ORIGINS, TOKENSAFE_AUTOPRUNE_INTERVAL, TOKENSAFE_BG_LICENSE_KEY, TOKENSAFE_BUNDLE_MAX_TOKENS, TOKENSAFE_CACHE_TIMEOUT, TOKENSAFE_CLOUD_DISTILL, TOKENSAFE_CLOUD_URL, TOKENSAFE_CONTEXT_WINDOW, TOKENSAFE_DATA_DIR, TOKENSAFE_DB_PATH, TOKENSAFE_GRAPH_DISK_CACHE, TOKENSAFE_GRAPH_TTL, TOKENSAFE_HMAC_SALT, TOKENSAFE_HOST, TOKENSAFE_LICENSE_KEY, TOKENSAFE_LIVE_FLUSH_SECONDS, TOKENSAFE_LIVE_URL, TOKENSAFE_MAX_EVENTS, TOKENSAFE_MAX_TURN_TOKENS, TOKENSAFE_MCP_AUTORELOAD, TOKENSAFE_MCP_TIMEOUT, TOKENSAFE_NONINTERACTIVE, TOKENSAFE_PRIVATE_KEY, TOKENSAFE_PROXY_DISABLED, TOKENSAFE_PUBLIC_KEY, TOKENSAFE_RETENTION_DAYS, TOKENSAFE_SELF_HOSTED, TOKENSAFE_STALE_GUARD, TOKENSAFE_TARGET_TURN_TOKENS, TOKENSAFE_TELEMETRY_DISABLED, TOKENSAFE_TEST_PROMPT, TOKENSAFE_TOOL_PRUNE_ALLOWLIST, TOKENSAFE_TOOL_PRUNE_KEEP, TOKENSAFE_TOOL_PRUNING, TOKENSAFE_TREESITTER only into requests to tokensafe.dev (and its subdomains) and refuses the rest.

$vet-skills run --secrets ANTHROPIC_API_KEY=tokensafe.dev,DEEPSEEK_API_KEY=tokensafe.dev,GEMINI_API_KEY=tokensafe.dev,OPENAI_API_KEY=tokensafe.dev,TOKENSAFE_AGENT=tokensafe.dev,TOKENSAFE_ALLOWED_ORIGINS=tokensafe.dev,TOKENSAFE_AUTOPRUNE_INTERVAL=tokensafe.dev,TOKENSAFE_BG_LICENSE_KEY=tokensafe.dev,TOKENSAFE_BUNDLE_MAX_TOKENS=tokensafe.dev,TOKENSAFE_CACHE_TIMEOUT=tokensafe.dev,TOKENSAFE_CLOUD_DISTILL=tokensafe.dev,TOKENSAFE_CLOUD_URL=tokensafe.dev,TOKENSAFE_CONTEXT_WINDOW=tokensafe.dev,TOKENSAFE_DATA_DIR=tokensafe.dev,TOKENSAFE_DB_PATH=tokensafe.dev,TOKENSAFE_GRAPH_DISK_CACHE=tokensafe.dev,TOKENSAFE_GRAPH_TTL=tokensafe.dev,TOKENSAFE_HMAC_SALT=tokensafe.dev,TOKENSAFE_HOST=tokensafe.dev,TOKENSAFE_LICENSE_KEY=tokensafe.dev,TOKENSAFE_LIVE_FLUSH_SECONDS=tokensafe.dev,TOKENSAFE_LIVE_URL=tokensafe.dev,TOKENSAFE_MAX_EVENTS=tokensafe.dev,TOKENSAFE_MAX_TURN_TOKENS=tokensafe.dev,TOKENSAFE_MCP_AUTORELOAD=tokensafe.dev,TOKENSAFE_MCP_TIMEOUT=tokensafe.dev,TOKENSAFE_NONINTERACTIVE=tokensafe.dev,TOKENSAFE_PRIVATE_KEY=tokensafe.dev,TOKENSAFE_PROXY_DISABLED=tokensafe.dev,TOKENSAFE_PUBLIC_KEY=tokensafe.dev,TOKENSAFE_RETENTION_DAYS=tokensafe.dev,TOKENSAFE_SELF_HOSTED=tokensafe.dev,TOKENSAFE_STALE_GUARD=tokensafe.dev,TOKENSAFE_TARGET_TURN_TOKENS=tokensafe.dev,TOKENSAFE_TELEMETRY_DISABLED=tokensafe.dev,TOKENSAFE_TEST_PROMPT=tokensafe.dev,TOKENSAFE_TOOL_PRUNE_ALLOWLIST=tokensafe.dev,TOKENSAFE_TOOL_PRUNE_KEEP=tokensafe.dev,TOKENSAFE_TOOL_PRUNING=tokensafe.dev,TOKENSAFE_TREESITTER=tokensafe.dev -- npx -y tokensafe

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

What changed since 3.1.5

vet-skills diff tokensafe 3.1.5 4.3.1

This version reaches for something 3.1.5 did not: 1 new host (z.ai) · 4 new credential variables (TOKENSAFE_CLOUD_DISTILL, TOKENSAFE_PROXY_DISABLED, TOKENSAFE_STALE_GUARD, TOKENSAFE_TEST_PROMPT). Compared from the two recorded runs, not from source.

New hosts
z.ai
New credential variables
TOKENSAFE_CLOUD_DISTILL, TOKENSAFE_PROXY_DISABLED, TOKENSAFE_STALE_GUARD, TOKENSAFE_TEST_PROMPT

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-20 and scored it 52/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; Destructive file system command detected; Access to sensitive file paths. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 21 s · 2 scenarios
t+CategoryObserved action
0.0sprocessno install scripts in bundle
20sfilesystemread ~/.tokensafe/venv/pyvenv.cfg · ~/.tokensafe/venv/bin/python3._pth · ~/.tokensafe/venv/lib/python3.11/site-packages · ~/.tokensafe/venv outside project
20sfilesystemwrote ~/.tokensafe/venv/pyvenv.cfg
20sclaimedstartup only: exited (0) before answering · ⚡ Installing TokenSafe dependencies (cryptography & tree-sitter AST grammars)... ❌ TokenSafe requires Python 3.10+ and 'cryptography>=42.0'. Install dependencies manually via: pip install 'cryptography>=42.0' 'tree-sitter>=0.23' 'tree-sitter-python'
20sprocessscript backend runs setup only
20sprocessno install scripts to re-run

Static findings

HIGH
Access to sensitive file pathssrc/secret_redactor.py:122 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesr"""process\.env(?:\.([A-Za-z0-9_]+)|\[\s*['"]([A-Za-z0-9_]+)['"]\s*\])\s*(?:\|\||\?\?)\s*['"`]([^'"`\n]+)['"`]""",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Destructive file system command detectedsrc/adapter_core.py:59 · Destructive file system command detectedr"rm\s+(-[a-zA-Z]*\s+)*-?[rf]{2,}[a-zA-Z]*\s+(--\s+)?/(?:\s|$)", # rm -rf /Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:465 · Python subprocess execution detectedproc = subprocess.run(["git", "grep", "-n", pattern, "--", search_dir],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:497 · Python subprocess execution detectedcount = subprocess.run(["git", "rev-list", "--count", "HEAD"], cwd=root,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathssrc/guard.py:8 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files5. Git Staging Protection for Sensitive Files (.env*, *.key, *.pem)Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:200 · Python subprocess execution detectedsubprocess.run([sys.executable, "-m", "venv", venv_dir],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathssrc/bench_scenarios.json:318 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files… }, "secret_guard": { "note": "Security scenario: a pre-registered FAKE .env fixture is written into the temp clone (values are synthetic, never real secret…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssrc/scoped_search.py:47 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"dockerfile", "dockerfile.cloud", "docker-compose.yml", ".env.example",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:504 · Python subprocess execution detectedproc = subprocess.run(["git", "diff", "HEAD~1", "HEAD"], cwd=root,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
CRITICAL
Destructive command against a filesystem root or devicesrc/adapter_core.py:59 · Destructive command against a filesystem root or devicer"rm\s+(-[a-zA-Z]*\s+)*-?[rf]{2,}[a-zA-Z]*\s+(--\s+)?/(?:\s|$)", # rm -rf /Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssrc/mcp_server.py:357 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesIgnores heavy folders (node_modules, dist, .git, etc.) and redacts sensitive files (.env*).Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Coercive compliance languagerules/AGENTS.md:8 · Coercive compliance languageWhen interacting within this environment, you MUST follow strict token economy guidelines:Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:334 · Python subprocess execution detectedprobe = subprocess.run([py, "-c", "import pytest_asyncio"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Destructive file system command detectedsrc/guard.py:275 · Destructive file system command detected# 1. Destructive Unbounded File Deletion (rm -rf /, rm -rf ~, rm -rf *, etc.)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssrc/guard.py:42 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files".env", ".env.local", ".env.production", ".env.staging", ".env.development",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:163 · Python subprocess execution detectedsubprocess.run(["git", "clone", "--depth", depth, "-q", repo["url"], dest],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Stripe-like API key pattern detectedsrc/bench_scenarios.json:319 · Stripe-like API key pattern detected — possible hardcoded or harvested key…)\nDATABASE_URL=postgres://admin:[email protected]:5432/prod\nAPI_KEY=sk-live-4c9a1f77e2b34d08a91c6d5e7f0b2a3d\nJWT_SECRET=9f3b1c77e2b34d08a91c6d5e7f0b2a3d9f3b\nSTRIPE_WEBHOOK_SECRET=whsec_5…Fix: Never read or enumerate credential locations; accept a documented environment variable instead.
AST03
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:208 · Python subprocess execution detectedsubprocess.run([venv_py, "-m", "pip", "install", "-q", "pytest", *deps, "-e", repo_root],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
CRITICAL
HTML comment carrying imperative instructions to the agentrules/AGENTS.md:1 · HTML comment carrying imperative instructions to the agentRule: Token-Optimized Communication Protocol (TOCP) Trigger: always_onFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:340 · Python subprocess execution detectedproc = subprocess.run(py_args, cwd=root, capture_output=True, text=True, encoding="utf-8", errors="replace", timeout=300)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:159 · Python subprocess execution detectedsubprocess.run(["git", "reset", "--hard", "-q", "origin/HEAD"], cwd=dest,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Access to sensitive file pathssrc/mcp_server.py:370 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files'.env', '.env.local', '.env.production', '.env.staging', '.env.development'Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssrc/benchmark_real.py:683 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesenv_path = os.path.join(root, ".env")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
HIGH
Access to sensitive file pathssrc/mcp_server.py:233 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files… without permissions, timestamps, or ignored folders (node_modules, .git, dist, .env). Replaces verbose ls -la. Pass '.' for project root or a subfolder like 'src'.…Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
CRITICAL
Pipe-to-shell patternsrc/installer.py:1094 · Pipe-to-shell pattern — downloads and immediately executes remote codeFeatures a resilient /dev/tty fallback for piped 'curl | bash' installations and anFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:157 · Python subprocess execution detectedsubprocess.run(["git", "fetch", "--depth", depth, "origin"], cwd=dest,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Destructive file system command detectedsrc/guard.py:9 · Destructive file system command detected6. Destructive Shell Command & Unbounded Deletion Prevention (rm -rf /)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
HIGH
Access to sensitive file pathssrc/compactor.py:271 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesr'|(?:\.env(?:\.[\w\-]+)?|\.gitignore|\.dockerignore|\.prettierrc[\w\.\-]*|\.eslintrc[\w\.\-]*|\.npmignore|\.npmrc|\.yarnrc)'Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:229 · Python subprocess execution detectedproc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
HIGH
Privileged sudo commandbin/tokensafe.js:150 · Privileged sudo command — agent may execute with elevated privilegesconsole.error(' Debian: sudo apt-get install python3\n');Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsrc/telemetry_sync.py:560 · Python subprocess execution detectedsubprocess.Popen(Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST05
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:262 · Python subprocess execution detectedsubprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/cli.py:1171 · Python subprocess execution detectedproc = subprocess.run(sub_cmd, shell=True, capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/image_downscale.py:174 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2477 · Python subprocess execution detectedres = subprocess.run([pip_cmd, "list"], capture_output=True, text=True, timeout=15)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:930 · Python subprocess execution detectedres = subprocess.run(["launchctl", "load", "-w", plist_path], capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:973 · Python subprocess execution detectedsubprocess.run(["systemctl", "--user", "disable", "--now", cls._SERVICE_NAME], capture_output=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2442 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:966 · Python subprocess execution detectedsubprocess.run(["launchctl", "unload", plist_path], capture_output=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2480 · Python subprocess execution detectedup_res = subprocess.run([pip_cmd, "install", "--upgrade", "tokensafe"], capture_output=True, text=True, timeout=120)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:288 · Python subprocess execution detectedsubprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2253 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2420 · Python subprocess execution detectedres = subprocess.run(["git", "pull", "--quiet"], cwd=PROJECT_ROOT, capture_output=True, text=True, timeout=10)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/guard.py:615 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:1061 · Python subprocess execution detectedproc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/benchmark_real.py:730 · Python subprocess execution detectedcapped = subprocess.run(rewritten, shell=True, cwd=root,Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/cli.py:351 · Python subprocess execution detectedsubprocess.run(["open", out_path], check=False)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/cli.py:353 · Python subprocess execution detectedos.startfile(out_path) # type: ignore[attr-defined]Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:929 · Python subprocess execution detectedsubprocess.run(["launchctl", "unload", plist_path], capture_output=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:954 · Python subprocess execution detectedres = subprocess.run(cmd, capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/mcp_proxy.py:99 · Python subprocess execution detectedproc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2452 · Python subprocess execution detectedup_res = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/license_manager.py:117 · Python subprocess execution detectedres = subprocess.run(cmd, capture_output=True, text=True, timeout=2)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/cli.py:1246 · Python subprocess execution detectedproc = subprocess.run(cmd, cwd=PROJECT_ROOT, capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:2525 · Python subprocess execution detectedsubprocess.run(sync_cmd, env=sync_env, capture_output=True, text=True, timeout=5)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/cli.py:355 · Python subprocess execution detectedsubprocess.run(["xdg-open", out_path], check=False)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/image_downscale.py:57 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:308 · Python subprocess execution detectedp = subprocess.Popen(zenity_cmd, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/installer.py:975 · Python subprocess execution detectedsubprocess.run(["systemctl", "--user", "daemon-reload"], capture_output=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/image_downscale.py:188 · Python subprocess execution detectedres = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Unpinned dependencysrc/installer.py:2451 · Unpinned dependencyprint(f" {CYAN}→{RESET} Updating global npm package via {npm_bin} (npm install -g tokensafe@latest)...")Fix: Pin the exact version (`[email protected]`, `==1.2.3`) so every install gets the bytes that were reviewed.
AST06
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:320 · Python subprocess execution detectedsubprocess.Popen(term)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Dynamic module loadingsrc/benchmark_real.py:311 · Dynamic module loadingif importlib.util.find_spec("pytest") is None:Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
npx invocation of non-well-known packagesrc/command_sanitizer.py:660 · npx invocation of non-well-known package — may execute untrusted code# 3. Linter commands (eslint, npx eslint, biome, npx biome)Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
File type combined with upload actionsrc/installer.py:404 · File type combined with upload action — agent may exfiltrate local data filesreturn True, "ZCode hooks registered in ~/.zcode/cli/config.json (SessionStart + UserPromptSubmit + PreToolUse guard + PostToolUse compact + Stop) — restart the app to take effect"Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
Remote installer piped to shellsrc/installer.py:1094 · Remote installer piped to shellFeatures a resilient /dev/tty fallback for piped 'curl | bash' installations and anFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:450 · Python subprocess execution detectedsubprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Fake instruction boundary tagREADME.md:456 · Fake instruction boundary tag · informational: not an instruction filetokensafe test "<prompt>" Preview prompt compaction on custom inputFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Python subprocess execution detectedsrc/version_checker.py:331 · Python subprocess execution detectedsubprocess.Popen([Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Instructs the agent to modify its persistent memory or instruction filessrc/environment_auditor.py:607 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction filelines.append(" \033[36m💡 Tip: Consolidate redundant rule files into a single AGENTS.md\033[0m")Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
npx invocation of non-well-known packagebin/tokensafe.js:5 · npx invocation of non-well-known package — may execute untrusted code* Works with `npx tokensafe [command]` or global `tokensafe [command]`Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
npx invocation of non-well-known packagesrc/command_sanitizer.py:640 · npx invocation of non-well-known package — may execute untrusted code# 2. TypeScript compiler & type-checking commands (tsc, npx tsc, npm run typecheck)Fix: Pin versions and sources; vendor anything fetched at install time.
AST06

Cost to the agent

Instructions loaded
≈2,339 tok
2 files · 81 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

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 tokensafe@4.3.1 as of 40 min ago. This is the package itself, not its dependencies.

What would make this Verified

28 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Reads credential files≈ +27 on the totalThe sandbox saw reads under ~/.ssh, ~/.aws, ~/.npmrc, ~/.config or similar. Read only files inside the project, and only the ones the task needs. Seen: ~/.tokensafe/venv/pyvenv.cfg, ~/.tokensafe/venv/bin/python3._pth, ~/.tokensafe/venv/lib/python3.11/site-packages.
  2. Destructive command against a filesystem root or device≈ +9 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  3. Pipe-to-shell pattern≈ +9 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  4. Unclaimed publisher≈ +8 on the totalClaim tokensafe-dev: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  5. Touches files outside the project≈ +7 on the totalKeep reads and writes inside the working directory, or declare the exact paths in `filesystem:` and explain why in the README.
  6. 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.
Publish a new version to npm, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim tokensafe-devPropose a patch to the maintainer