Trust Assessment
gong received a trust score of 25/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 4 findings: 1 critical, 3 high, 0 medium, and 0 low severity. Key findings include Hidden network beacons / undisclosed telemetry, Command Injection via unsanitized date arguments, JSON Injection in API request body.
The analysis covered 4 layers: manifest_analysis, llm_behavioral_safety, static_code_analysis, dependency_graph. The static_code_analysis layer scored lowest at 55/100, indicating areas for improvement.
Last analyzed on February 11, 2026 (commit 0676c56a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Command Injection via unsanitized date arguments The `calls` and `stats` commands in `gong.sh` construct date strings using a user-provided number of days (`$DAYS`). This variable is directly interpolated into the `date` command (specifically the BSD `date -v` variant). If `$DAYS` contains shell metacharacters (e.g., `;`, `|`, `&`, `$(...)`), an attacker can inject and execute arbitrary shell commands, leading to remote code execution on systems where BSD `date` is used (e.g., macOS). Sanitize the `$DAYS` variable to ensure it only contains digits, or use a safer method to calculate dates that does not involve direct shell interpolation of user input. For example, validate `$DAYS` is an integer using a regex or `[[ $DAYS =~ ^[0-9]+$ ]]` before using it in the `date` command. | Unknown | scripts/gong.sh:30 | |
| HIGH | Hidden network beacons / undisclosed telemetry Command output piped through base64 encoding Remove undisclosed network calls and telemetry. All outbound communication should be documented and necessary for the skill's stated purpose. BCC injection in email tools is almost always malicious. | Unknown | /var/folders/1k/67b8r20n777f_xcmmm8b7m5h0000gn/T/skillscan-clone-5dwmslzu/repo/skills/gong/SKILL.md:24 | |
| HIGH | Hidden network beacons / undisclosed telemetry Command output piped through base64 encoding Remove undisclosed network calls and telemetry. All outbound communication should be documented and necessary for the skill's stated purpose. BCC injection in email tools is almost always malicious. | Unknown | /var/folders/1k/67b8r20n777f_xcmmm8b7m5h0000gn/T/skillscan-clone-5dwmslzu/repo/skills/gong/scripts/gong.sh:16 | |
| HIGH | JSON Injection in API request body The `call` and `transcript` commands directly interpolate the user-provided `call_id` (`$1`) into the JSON request body for the `-d` argument of `curl` without proper escaping. An attacker could inject JSON syntax (e.g., `"` or `\`) to alter the API request's structure. While this might lead to malformed JSON and API errors, a sophisticated attacker could potentially manipulate the API request parameters in unintended ways, leading to data exposure, unauthorized actions, or denial of service if the API is permissive or handles malformed JSON in an exploitable manner. Properly escape user-provided input before embedding it into JSON strings. A robust solution involves using a tool like `jq` to construct the JSON payload, ensuring all special characters are correctly escaped. For example, `jq -n --arg id "$1" '{"filter":{"callIds":[$id]},"contentSelector":{"exposedFields":{"content":true,"parties":true}}}'`. | Unknown | scripts/gong.sh:40 |
Scan History
Embed Code
[](https://skillshield.io/report/6517443ed8753ff4)
Powered by SkillShield