The Agent Skills Directory

[SAFE]: The skill operates entirely on local files such as manuscript drafts, figures, and tables. No external domains or network communication patterns were detected.
[SAFE]: The instructions focus on narrative restructuring and scientific framing, adhering to legitimate academic standards and professional workflows.
[SAFE]: The skill includes safety constraints (Stop Conditions) that prevent the AI from proceeding if critical data is missing or if it is asked to make unsupported scientific claims.
[PROMPT_INJECTION]: The skill processes untrusted external data (manuscript drafts and supplementary files) which creates an inherent attack surface for indirect prompt injection.
Ingestion points: Manuscript source paths, figures, tables, and result summaries (SKILL.md).
Boundary markers: No explicit delimiters are used to separate the processed file content from the skill's core instructions.
Capability inventory: The skill has access to Bash, Edit, and Write tools for file system interaction.
Sanitization: No data validation or sanitization of input files is specified.

manuscript-reframe