The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use a custom wiki CLI and bun for various maintenance tasks. Specifically, it suggests running bun run sync:local to synchronize edited skills, which involves shell-level script execution.
[EXTERNAL_DOWNLOADS]: The wiki source ingest <path-or-url> command allows the agent to download content from arbitrary remote locations. This is a primary feature for research filing but necessitates the handling of potentially malicious external data.
[DATA_EXFILTRATION]: Because the ingestion command accepts both local file paths and remote URLs, there is a risk that an agent could be induced to read sensitive local files and transmit their contents or include them in research logs that are later exported.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to ingest and process data from external, untrusted sources.
Ingestion points: Untrusted data enters the context via wiki research ingest, wiki source ingest <path-or-url>, and by reading git history or filesystem globs.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when the agent processes the ingested content.
Capability inventory: The agent has the capability to write to the filesystem, perform network operations, and execute CLI commands based on its analysis.
Sanitization: There is no evidence of content sanitization or validation before the data is processed or used to influence agent decisions.

wiki