The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access behaviors were detected. The skill is limited to text analysis and local file creation.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from the user conversation.
Ingestion points: Current conversation history (SKILL.md).
Boundary markers: Absent; instructions do not explicitly tell the agent to ignore instructions embedded in the analyzed text.
Capability inventory: Ability to write and update a local file (UBIQUITOUS_LANGUAGE.md).
Sanitization: Absent; the skill does not specify validation or filtering of extracted terms.
[DATA_EXPOSURE]: The skill reads conversation history and writes to a local file system path (UBIQUITOUS_LANGUAGE.md). This is the intended behavior for terminology management and does not involve sensitive system files or credentials.

ubiquitous-language