Security audit
Security checks for vulnerabilities and agentic risk
This is a real NYC event-search skill, but it asks the assistant to retain user account tokens and send personalized context to SyncSo with broad activation rules.
Review this before installing if you do not want a third-party event service receiving NYC plans, preferences, location, or calendar-derived context. Only provide a SyncSo code if you are comfortable with the assistant using your account allowance and retaining that token; prefer explicit confirmation before searches, calendar use, standing daily checks, or storing credentials.
The skill description uses very broad activation cues such as 'what to do,' 'where to go,' and 'where to take people in New York,' which can match many ordinary conversations and cause over-invocation. That can redirect unrelated user requests into this third-party skill and unnecessarily send user context to an external service.
The main guidance repeats expansive trigger conditions and instructs the agent to use the search tool whenever the user asks broadly 'what to do' or 'want a plan,' without meaningful limits. In context, this increases the chance of unnecessary tool use and external data disclosure because the skill pushes itself into many borderline situations.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
**If it does not**, the tool definitions are at the end of this file.
Add them to your model's tools and POST each call as JSON-RPC to that same
URL, carrying the access token from the section below. The reply is
`result.content[0].text` — compact text, about 120 tokens per result, ready
to hand straight back to the model.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
**If it does not**, the tool definitions are at the end of this file.
Add them to your model's tools and POST each call as JSON-RPC to that same
URL, carrying the access token from the section below. The reply is
`result.content[0].text` — compact text, about 120 tokens per result, ready
to hand straight back to the model.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
fetch it again rather than answer from stale guidance.
curl -sS https://rtdb.syncso.com/partner/mcp \
-H "Authorization: Bearer $SYNCSO_TOKEN" \
-H "X-SyncSo-Skill: 1.0.1" \
-H "Content-Type: application/json" \
No suspicious patterns detected.