Trade Signal
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill’s market-data API use is disclosed, but its search script handles the query unsafely and could run local code if given a specially crafted question.
Review carefully before installing or running. The external API call is expected for this skill, but the helper script should be fixed so user questions are passed as data rather than embedded into Python code. Treat outputs as research support, not automatic financial advice.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
57/57 vendors flagged this skill as clean.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A maliciously crafted prompt or copied market text could cause the helper script to run local code with the user's permissions when the agent invokes it.
The unescaped QUERY value is embedded directly inside Python source passed to python3 -c; a query containing triple single quotes can break out of the literal and execute arbitrary Python code.
ENCODED=$(python3 -c "import urllib.parse; print(urllib.parse.quote('''$QUERY'''))")Fix the script to pass the query as data, not source code, for example: python3 -c 'import urllib.parse,sys; print(urllib.parse.quote(sys.argv[1]))' "$QUERY" or curl --get --data-urlencode "query=$QUERY". Avoid running it on untrusted query text until fixed.
Terminal-X can receive and process the questions sent through the skill, including any sensitive financial context the user includes.
The helper sends the user's encoded trading question to an external Terminal-X API endpoint. This is disclosed and purpose-aligned, but the query may contain portfolio or investment details.
curl -sL "https://app.terminal-x.ai/api/lite-search?query=${ENCODED}"Do not include confidential account numbers, full portfolio details, or other sensitive personal information unless you trust the provider and its privacy practices.
Users may have less clarity about provenance, version alignment, and required local tools before running the skill.
The registry metadata does not identify the source and does not declare the local tools used by the included script; SKILL.md also shows a different version. This is a transparency issue, not proof of malicious behavior.
Source: unknown; Version: 1.0.6; Required binaries (all must exist): none; Install specifications: No install spec
Publish or link the source, align the registry and SKILL.md versions, and declare bash/python3/curl requirements.