Trust Assessment
clawsend received a trust score of 10/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 20 findings: 2 critical, 2 high, 15 medium, and 1 low severity. Key findings include Arbitrary command execution, Unsafe deserialization / dynamic eval, Suspicious import: requests.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Manifest Analysis layer scored lowest at 21/100, indicating areas for improvement.
Last analyzed on February 12, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings20
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Python shell execution (os.system, subprocess) Review all shell execution calls. Ensure commands are static (not built from user input), use absolute paths, and are strictly necessary. Prefer library APIs over shell commands. | Manifest | skills/tlxue/clawsend/python/scripts/receive.py:44 | |
| CRITICAL | Command Injection via --on-message callback The `receive.py` script uses `subprocess.run` with `shell=True` to execute a command provided via the `--on-message` argument. If an attacker can control the `--on-message` argument (e.g., through prompt injection to the LLM that invokes this script), they can execute arbitrary shell commands on the host system. The message content is passed to the command via stdin, but the command itself is directly executed by the shell. Avoid using `shell=True` when executing external commands with untrusted input. Instead, parse the command and its arguments, then pass them as a list to `subprocess.run` without `shell=True`. If `shell=True` is absolutely necessary, implement strict validation and sanitization of the `command` argument to ensure it only contains allowed executables and arguments. | LLM | python/scripts/receive.py:61 | |
| HIGH | Dangerous call: subprocess.run() Call to 'subprocess.run()' detected in function 'execute_callback'. This can execute arbitrary code. Avoid using dangerous functions like exec/eval/os.system. Use safer alternatives. | Static | skills/tlxue/clawsend/python/scripts/receive.py:44 | |
| HIGH | Arbitrary Directory Deletion via --vault-dir and --force The `generate_identity.py` script allows specifying a custom vault directory via `--vault-dir`. If the `--force` flag is also used, the script will recursively delete the specified `vault.vault_dir` using `shutil.rmtree`. If an attacker can control the `--vault-dir` argument, they could potentially delete arbitrary directories on the file system, leading to data loss or denial of service. Restrict the `--vault-dir` argument to a safe, controlled location (e.g., within the skill's own dedicated data directory) and prevent it from being set to arbitrary paths by untrusted input. Implement strict validation for `vault_dir` to ensure it does not escape expected boundaries. Consider requiring explicit user confirmation for `--force` operations, especially when `--vault-dir` is custom. | LLM | python/scripts/generate_identity.py:51 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/node/lib/crypto.js:4 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/node/lib/crypto.js:229 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/node/scripts/receive.js:44 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/python/lib/client.py:4 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/python/lib/crypto.py:7 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/python/lib/envelope.py:5 | |
| MEDIUM | Unsafe deserialization / dynamic eval Decryption followed by code execution Remove obfuscated code execution patterns. Legitimate code does not need base64-encoded payloads executed via eval, encrypted-then-executed blobs, or dynamic attribute resolution to call system functions. | Manifest | skills/tlxue/clawsend/python/scripts/heartbeat.py:13 | |
| MEDIUM | Suspicious import: requests Import of 'requests' detected. This module provides network or low-level system access. Verify this import is necessary. Network and system modules in skill code may indicate data exfiltration. | Static | skills/tlxue/clawsend/python/lib/client.py:13 | |
| MEDIUM | Suspicious import: requests Import of 'requests' detected. This module provides network or low-level system access. Verify this import is necessary. Network and system modules in skill code may indicate data exfiltration. | Static | skills/tlxue/clawsend/python/scripts/discover.py:71 | |
| MEDIUM | Unpinned npm dependency version Dependency '@noble/curves' is not pinned to an exact version ('^1.3.0'). Pin dependencies to exact versions to reduce drift and supply-chain risk. | Dependencies | skills/tlxue/clawsend/node/package.json | |
| MEDIUM | Unpinned Python dependency version Requirement 'flask>=3.0' is not pinned to an exact version. Pin Python dependencies with '==<exact version>'. | Dependencies | skills/tlxue/clawsend/python/requirements.txt:4 | |
| MEDIUM | Unpinned Python dependency version Requirement 'gunicorn>=21.0' is not pinned to an exact version. Pin Python dependencies with '==<exact version>'. | Dependencies | skills/tlxue/clawsend/python/requirements.txt:7 | |
| MEDIUM | Unpinned Python dependency version Requirement 'cryptography>=42.0' is not pinned to an exact version. Pin Python dependencies with '==<exact version>'. | Dependencies | skills/tlxue/clawsend/python/requirements.txt:10 | |
| MEDIUM | Unpinned Python dependency version Requirement 'requests>=2.31' is not pinned to an exact version. Pin Python dependencies with '==<exact version>'. | Dependencies | skills/tlxue/clawsend/python/requirements.txt:13 | |
| MEDIUM | Unpinned Python Dependencies The `python/requirements.txt` file specifies dependencies using loose version constraints (e.g., `flask>=3.0`). This allows `pip` to install any future version that meets the minimum requirement, including potentially vulnerable or malicious versions, without explicit review. This increases the risk of supply chain attacks. Pin all dependencies to exact versions (e.g., `flask==3.0.0`). Use a dependency lock file (e.g., `pip freeze > requirements.lock`) and ensure it is used for all installations. Regularly audit and update dependencies to known good versions. | LLM | python/requirements.txt:1 | |
| LOW | Unpinned Node.js Dependencies The `node/package.json` file specifies dependencies using caret (`^`) version ranges (e.g., `@noble/curves: ^1.3.0`). While `package-lock.json` provides exact versions for reproducible builds, a fresh `npm install` without a lock file could pull in vulnerable minor versions if a malicious update is published. This introduces a potential, albeit mitigated, supply chain risk. Consider pinning exact versions in `package.json` or ensure that `package-lock.json` is always present and used for all deployments to guarantee reproducible and secure dependency installations. Regularly audit and update dependencies. | LLM | node/package.json:17 |
Scan History
Embed Code
[](https://skillshield.io/report/00a1d7c1a8290633)
Powered by SkillShield