The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script scripts/query.js uses require() to load a module from a path computed at runtime (distPath). This module is expected to be at a location relative to the script's directory. Dynamic loading from computed paths is a security risk as it can be exploited to execute arbitrary code if the filesystem environment or the path variables are manipulated.\n- [INDIRECT_PROMPT_INJECTION]: This skill processes content from a wiki database and provides it to the AI agent's context, creating a surface for indirect attacks.\n
Ingestion points: Data enters the context via the store.searchWiki and store.getWikiPage functions in scripts/query.js.\n
Boundary markers: There are no boundary markers or instructions to isolate the retrieved wiki content, increasing the risk that the agent interprets data as instructions.\n
Capability inventory: The skill possesses the ability to execute shell commands and access the filesystem via Node.js.\n
Sanitization: The script performs formatting for display but does not include any sanitization to detect or neutralize prompt injection patterns within the wiki data.

wiki-query