Security audit
Security checks for vulnerabilities and agentic risk
This is a non-executable planning skill for a specific housing-dispute help request, with the main risk being overly broad automatic triggers.
Install only if you want an implicit helper for this specific housing-dispute topic. The publisher should narrow the activation terms or require explicit invocation, because generic words like 'ask' and 'hello' may cause the skill to appear in unrelated requests.
The skill metadata description says it should activate for broad terms like 'general-help', 'ask-hn', 'ask', and 'hello'. Terms such as 'ask' and 'hello' are common in ordinary conversations, so the skill may be invoked outside its intended housing-dispute context, causing context hijacking or misrouting of user requests.
The trigger keyword list includes highly ambiguous terms including 'ask', 'hello', and 'landing' without any contextual constraints. Because these words are common across unrelated requests, they can cause unintended invocation of this skill, leading to incorrect assistance, suppression of more appropriate skills, or attacker-crafted prompts that deliberately force skill selection.
The usage signals are broad enough to match generic terms like 'ask', 'general-help', and fragments of the topic, which can cause this skill to be invoked for unrelated requests. Unintended invocation can route users into an inappropriate workflow, causing misleading guidance, context confusion, and potential interference with more suitable skills.
No suspicious patterns detected.
