Back to skill

Security audit

provian

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because the scanners flag a token-bearing request that follows redirects and an install command that uses an unpinned package version.

Before installing, consider whether you trust the service to receive the connection token. The scanner flags possible token exposure through redirects and a package command that can select a changing version; these are security concerns, not evidence that the author acted maliciously. The scans inspected both reported components, though the AIG scan returned no findings; a clean result from that scanner does not remove the identified concerns.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
compatibility: >-
  Runtime-agnostic. Path A needs curl and a POSIX shell (macOS, Linux) or PowerShell
  (Windows). Path B needs Node.js 20 or later with npx. Path C needs only an MCP
  client that can reach https://api.provian.ai/mcp. Every path needs network access
  to api.provian.ai.
metadata:
  homepage: https://provian.ai/agents

MCP server referenced without pinned version: 'npx envVars'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
94% confidence
Finding

Sending the connection token to Provian is expected, but curl -L follows redirects while the token is supplied as a custom x-provian-token header. Such custom headers can be forwarded to a different redirect destination, exposing the one-time connection credential if the endpoint issues an attacker-controlled redirect; using stdin protects process arguments but not this transmission path.

Content

Scanner excerpt · SKILL.md (reported line 108)May include surrounding context.

sh
umask 077
curl -fsSL -H @- -o provian-install.sh "https://api.provian.ai/v1/bootstrap/agent" <<EOF
x-provian-token: ${PROVIAN_CONNECTION_TOKEN}
EOF
cat provian-install.sh     # read what it does before running it

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 120)May include surrounding context.

powershell
$headers = @{ 'x-provian-token' = $env:PROVIAN_CONNECTION_TOKEN; 'x-provian-platform' = 'windows' }
Invoke-RestMethod -Headers $headers -Uri 'https://api.provian.ai/v1/bootstrap/agent' -OutFile provian-install.ps1
Get-Content provian-install.ps1    # read what it does before running it
& .\provian-install.ps1    # execute the local file only after reviewing it
Remove-Item provian-install.ps1

MCP server referenced without pinned version: 'npx @provian/agent-sdk'.

Medium
Category
MCP Rug Pull
Confidence
98% confidence
Finding

The recommended command executes an npm package selected by the mutable latest tag rather than a reviewed, fixed release. A compromised package publication could execute with the agent's permissions, access its connection token and wallet material, or install malicious persistent components; optional pinning advice does not secure the default command.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 151)May include surrounding context.

md
Add Provian's MCP server to your MCP client:

- URL: `https://api.provian.ai/mcp` (streamable HTTP)
- Credential: none yet. After step 3 you put the returned credential in the
  connector's API-key field.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 231)May include surrounding context.

md
- Agents and developers: https://provian.ai/agents and https://provian.ai/developers
- Docs written for agents to read: https://provian.ai/llms.txt
- MCP descriptor: https://api.provian.ai/.well-known/mcp.json
- Contact: hello@provian.ai

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 12)May include surrounding context.

text
permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE