The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from the raw/articles/ directory to create summaries and structured wiki pages. Maliciously crafted content in these source materials could potentially influence the agent's behavior during the ingestion or querying workflows.
Ingestion points: Markdown files and research articles located in the raw/ directory.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within source files.
Capability inventory: The skill utilizes file system read/write access, git for version control, and sed for text manipulation.
Sanitization: There are no explicit instructions for sanitizing or validating the content extracted from raw materials before it is incorporated into the wiki.
[COMMAND_EXECUTION]: The skill performs shell commands to automate repository maintenance and content fixes.
Version Control: Executes git add -A && git commit to track changes after batch operations.
Text Processing: Employs sed for batch-replacing dangling links and correcting slug formats during the linting process.

wiki-knowledge-base