The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface because it is designed to ingest and process user-provided research data and manuscript content which may contain malicious instructions.
Ingestion points: The skill processes user-supplied statistical results, data descriptions, and manuscript drafts to verify reporting norms (file: SKILL.md).
Boundary markers: No specific boundary markers or instructions to ignore embedded commands in the processed data are provided.
Capability inventory: The skill references an 'Execution bridge' that suggests executing complex statistical functions and analysis batteries via an external bridge (file: SKILL.md).
Sanitization: No data sanitization or input validation steps are mentioned for the processed data.
[COMMAND_EXECUTION]: The skill integrates with an external execution bridge for running statistical code.
Evidence: The skill references an 'Execution bridge' and specifies functions such as romano_wolf, benjamini_hochberg, and audit_result to be used for data analysis and reporting.
[EXTERNAL_DOWNLOADS]: The skill recommends several well-known third-party statistical packages.
Evidence: Mentions established tools like Mplus, lavaan, lme4, nlme, psych, metafor, and metaSEM for various statistical modeling tasks (file: SKILL.md).

joap-data-analysis