skills/brycewang-stanford/awesome-agent-skills-for-empirical-research/bayesian-estimation/Gen Agent Trust Hub
bayesian-estimation
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, hidden instructions, or unauthorized data access were identified. The skill focuses on legitimate statistical modeling workflows.\n- [COMMAND_EXECUTION]: The skill provides instructions for executing Python and R code for statistical analysis. These examples use established and well-regarded libraries such as PyMC, ArviZ, and Stan for MCMC sampling and convergence diagnostics.\n- [EXTERNAL_DOWNLOADS]: The skill references several well-known scientific libraries as dependencies, including
pymc,arviz,cmdstanpy,numpyro,jax,scipy, andnumpy. These are official, standard packages within the Bayesian statistical community.\n- [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface associated with the ingestion of literature benchmarks via external agents (e.g.,methods-explorer).\n - Ingestion points: Literature calibration targets referenced in
SKILL.md.\n - Boundary markers: Absent.\n
- Capability inventory: Python script execution, Stan compilation via
cmdstanpy.\n - Sanitization: Absent; the skill assumes data retrieved by external agents is trustworthy or will be reviewed by the user.
Audit Metadata