The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown instructions and a documentation asset. It does not include any scripts, executable code, or commands that interact with the network or the operating system.
[NO_CODE]: The skill lacks any functional code (Python, JavaScript, shell scripts) or package dependencies, relying solely on natural language instructions to guide the agent's behavior.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from 'raw sources' to answer questions. While this creates a surface for indirect prompt injection, the agent's capabilities are restricted to reading and writing within a specific wiki directory structure.
Ingestion points: Reads files from the wiki/ directory and external 'raw sources' as described in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters to isolate untrusted data, though they do require the agent to distinguish between 'extracted' and 'inferred' content.
Capability inventory: Limited to file read/write operations within the local project wiki; no network, subprocess, or shell command tools are used.
Sanitization: No input sanitization or content validation logic is present.

wiki-query