The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes filesystem read and write capabilities to manage wiki files within a user-defined directory.
Evidence: The skill performs extensive file operations (reading config/schema, writing new pages, updating index/log files) but includes a protective 'Scope check' in SKILL.md that mandates an immediate stop if the access scope is directed at system roots or user home directories.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and process untrusted text, such as chat transcripts or research notes.
Ingestion points: User-provided transcripts, research notes, or thread summaries (Workflow Step 2 in SKILL.md).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the processed data.
Capability inventory: Extensive filesystem read/write operations (Steps 4a, 4b, 5, 7 in SKILL.md).
Sanitization: There are no explicit instructions to sanitize or validate the content of the ingested data before it is written to the wiki files. However, this is a common characteristic of data-processing skills and is mitigated by the skill's highly structured workflow.

wiki-crystallize