Back to skill

Security audit

Sahaba Ijma

Security checks for vulnerabilities and agentic risk

Overview

This search skill mostly matches its stated purpose, but its query handling can turn a crafted search term into local code execution.

Install only after the query-encoding bug is fixed. Treat any search text as data sent to dorar.net, and avoid using untrusted or sensitive query strings until the shell script passes the query to Python as an argument or uses curl data-urlencode instead of building Python source dynamically.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/sahaba_search.sh:12
Finding

Arbitrary Python Code Execution Through Unsafely Interpolated Search Query

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · scripts/sahaba_search.sh (reported line 14)May include surrounding context.

sh
ENCODED=$(python3 -c "import urllib.parse; print(urllib.parse.quote('$QUERY'))")

RESULT=$(curl -s -L \
  -H "User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" \
  -H "Accept: application/json, text/javascript, */*" \
  -H "Referer: https://dorar.net/" \

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script sends the user's raw search query to dorar.net over the network without any user-facing notice or consent prompt. Even if this is necessary for functionality, queries may contain sensitive religious, personal, or investigative topics, creating a privacy leak to a third party.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

With no manifest available, this script appears to be a stdin JSON/HTML parsing utility, but it additionally reads a query from the process environment to construct and emit a search URL. Accessing environment-derived context is not obviously required for the parser behavior implemented elsewhere in the file, so this is an unjustified extra capability relative to the apparent purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The natural-language description states that the script returns results in Arabic, but it does not indicate that language selection is optional or justified as a region-specific constraint. This can be a locale policy concern because the skill appears to impose a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

Exporting the user query as an environment variable makes the value available to child processes, increasing accidental exposure through process inspection, crash reports, debugging, or downstream tooling. This is a smaller issue than network exfiltration, but it still broadens the query's exposure surface without disclosure.

Content

No source excerpt is available for this finding.