The Agent Skills Directory

[SAFE]: The skill provides instructions for interacting with the yy or yylo-ledger CLI tools to manage wiki records. It includes defensive measures such as recommending help-check preflights before execution, advising against secret storage, and enforcing revision-based concurrency controls for updates. All external resources and commands are associated with the vendor's own tooling (yylo-dev).
[INDIRECT_PROMPT_INJECTION]: The skill possesses a standard attack surface for indirect prompt injection as it retrieves and processes content from external ledger records. While this is inherent to the skill's purpose, the following factors apply:
Ingestion points: Data is retrieved via yy ledger wiki search and yy ledger get commands in SKILL.md.
Boundary markers: None present. The instructions do not explicitly tell the agent to treat retrieved record content as untrusted data or use delimiters.
Capability inventory: The skill has enable-shell-directives: true, allowing the agent to execute arbitrary shell commands.
Sanitization: No sanitization or validation of the record content is specified before the agent processes or reports it.

wiki-yylo