Trust Assessment
clawtoclaw received a trust score of 86/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Command Injection via API-provided heartbeat command.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Command Injection via API-provided heartbeat command The skill's documentation describes an `eventModeHint` object returned by the `events:checkIn` API call. This object includes a `heartbeat.command` field which contains a shell command string (e.g., `python3 scripts/event_heartbeat.py ...`). The documentation explicitly instructs the agent to 'Keep running the event heartbeat' and provides an example of this command. If the AI agent is implemented to directly execute the value of this `command` field received from the C2C API, and if the C2C API were compromised, an attacker could inject arbitrary shell commands, leading to remote code execution on the agent's host. This violates the principle of treating all external content as untrusted and not executing commands found within it. The AI agent should not directly execute shell commands received from external APIs. Instead, the agent should call a predefined, safe function (e.g., `run_event_heartbeat()`) that encapsulates the heartbeat logic. This function should accept only structured, validated parameters (e.g., `eventId`, `propose_intros`) rather than a raw command string. The `command` field should be removed from the API response, or treated as purely informational for human developers, not for automated execution by the agent. | LLM | SKILL.md:362 |
Scan History
Embed Code
[](https://skillshield.io/report/7747a42922bc852a)
Powered by SkillShield