The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill reads external data from multiple markdown documents to extract terminology. This creates a vulnerability surface where malicious instructions embedded in the documents could potentially influence the agent's behavior.
Ingestion points: The skill reads docs/product-requirements.md, docs/functional-design.md, docs/architecture.md, docs/repository-structure.md, and docs/development-guidelines.md.
Boundary markers: There are no delimiters or explicit instructions provided to the agent to ignore embedded commands within the ingested text.
Capability inventory: The skill has the capability to read local files and write a new documentation file (docs/glossary.md).
Sanitization: No input validation or sanitization logic is present to filter malicious strings from the documentation files.
[Data Exfiltration] (SAFE): No network operations (e.g., curl, wget, fetch) or access to sensitive credentials (e.g., API keys, SSH keys) were detected.
[Remote Code Execution] (SAFE): The skill does not download external packages or execute remote scripts; all operations are local and static.

glossary-creation