The Agent Skills Directory

[METADATA_POISONING]: The skill exhibits deceptive metadata concerning its authorship and file structure.
The YAML frontmatter lists "Orchestra Research" as the author, which conflicts with the identified author "hurtzdonut559". This discrepancy could mislead users regarding the official status or origin of the skill.
SKILL.md refers to a documentation file named references/llms-txt.md, but the file provided in the skill structure is references/llms.md. This internal inconsistency indicates an automated or improperly managed generation process.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it is designed to ingest and process documentation data to assist the agent.
Ingestion points: Reference files located in the references/ directory (llms.md and index.md).
Boundary markers: The instructions do not include specific delimiters or "ignore embedded instructions" warnings for the data ingested from the reference files.
Capability inventory: Analysis of all files reveals no dangerous capabilities such as subprocess calls, file-write operations, or network requests.
Sanitization: No escaping, validation, or filtering mechanisms are defined to handle potentially malicious instructions embedded within the documentation content.

unsloth