The command findings are false positives caused by Markdown fences and API examples, not executable code in the skill. The hardcoded SoulTrace endpoint is real network behavior, and the skill transmits personality answers to a third-party service. No prompt injection text was found.
The skill requires sending all accumulated personality answers to soultrace.app for each assessment request. These answers may reveal personal traits, but the skill gives no privacy, retention, or consent guidance.
The protocol explicitly says all accumulated answers are sent with every request. The skill is clearly a personality assessment, so the data has personal sensitivity.
Capability review items (5)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The sample response includes a SoulTrace hosted result URL, confirming third-party result access. This is expected behavior but remains an external network privacy consideration.
The skill documents a GET request to the fixed SoulTrace API endpoint. This is expected metadata access but still requires external network communication.
The command findings are false positives caused by Markdown fences and API examples, not executable code in the skill. The hardcoded SoulTrace endpoint is real network behavior, and the skill transmits personality answers to a third-party service. No prompt injection text was found.
The skill requires sending all accumulated personality answers to soultrace.app for each assessment request. These answers may reveal personal traits, but the skill gives no privacy, retention, or consent guidance.
The protocol explicitly says all accumulated answers are sent with every request. The skill is clearly a personality assessment, so the data has personal sensitivity.
Capability review items (5)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
The sample response includes a SoulTrace hosted result URL, confirming third-party result access. This is expected behavior but remains an external network privacy consideration.
The skill documents a GET request to the fixed SoulTrace API endpoint. This is expected metadata access but still requires external network communication.
Static command execution alerts are false positives from fenced markdown examples and inline code snippets. The real risk is network transmission of personality assessment answers to https://soultrace.app/api/agent, which should be disclosed before use. Weak cryptography and system reconnaissance alerts were not supported by the reviewed lines.
1
Files scanned
172
Lines analyzed
3
Review items
3
False positives ignored
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
TRUE_POSITIVE: The skill instructs agents to send user assessment answers to the SoulTrace API and to accumulate all prior answers with each request. These answers can reveal personality traits and should be treated as personal data sent to an external service.
The endpoint and workflow are explicitly documented, and the skill requires sending accumulated answers to the remote API. The risk is privacy exposure, not evidence of malicious exfiltration.
Static false positives ignored (3)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
FALSE_POSITIVE: The Ruby or shell backtick detections correspond to fenced markdown blocks, inline JSON examples, and API response examples. No executable script file, install hook, or hidden runtime command was found in the reviewed skill.
The referenced lines are markdown documentation and examples, not code that runs automatically. The skill can still cause network requests when an agent follows the documented API flow.
FALSE_POSITIVE: The flagged lines describe a psychological model and answer collection behavior. No MD5, SHA1, DES, RC4, or other weak cryptographic algorithm use was found on those lines.
Direct review of the flagged lines shows personality-assessment text, not cryptographic operations. The finding appears to be a keyword collision.
FALSE_POSITIVE: The flagged line is an HTTP error table entry for status 400. It does not request host, user, process, environment, or network interface information.
The line only documents an invalid request response. No system enumeration behavior is present in the reviewed file.
Static command execution alerts are false positives from fenced markdown examples and inline code snippets. The real risk is network transmission of personality assessment answers to https://soultrace.app/api/agent, which should be disclosed before use. Weak cryptography and system reconnaissance alerts were not supported by the reviewed lines.
1
Files scanned
172
Lines analyzed
3
Review items
3
False positives ignored
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
TRUE_POSITIVE: The skill instructs agents to send user assessment answers to the SoulTrace API and to accumulate all prior answers with each request. These answers can reveal personality traits and should be treated as personal data sent to an external service.
The endpoint and workflow are explicitly documented, and the skill requires sending accumulated answers to the remote API. The risk is privacy exposure, not evidence of malicious exfiltration.
Static false positives ignored (3)
These static matches were dismissed by semantic review or matched schema-only tokens, so they are shown for transparency but do not drive the quality score.
FALSE_POSITIVE: The Ruby or shell backtick detections correspond to fenced markdown blocks, inline JSON examples, and API response examples. No executable script file, install hook, or hidden runtime command was found in the reviewed skill.
The referenced lines are markdown documentation and examples, not code that runs automatically. The skill can still cause network requests when an agent follows the documented API flow.
FALSE_POSITIVE: The flagged lines describe a psychological model and answer collection behavior. No MD5, SHA1, DES, RC4, or other weak cryptographic algorithm use was found on those lines.
Direct review of the flagged lines shows personality-assessment text, not cryptographic operations. The finding appears to be a keyword collision.
FALSE_POSITIVE: The flagged line is an HTTP error table entry for status 400. It does not request host, user, process, environment, or network interface information.
The line only documents an invalid request response. No system enumeration behavior is present in the reviewed file.
Static scanner detected 21 potential issues but all are false positives. The 13 external_commands findings are curl examples in Markdown documentation (not executable code). The 2 blocker findings misidentified Bayesian statistical inference as weak cryptography. The 5 network findings are legitimate API calls to soultrace.app, the skill's own service. Only risk is external network dependency which is expected for API integration skills.
1
Files scanned
172
Lines analyzed
2
Review items
0
False positives ignored
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Skill makes HTTP requests to external API (soultrace.app) for personality assessment functionality. This is expected behavior for API integration but introduces network dependency.