The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides functional Python scripts (scripts/agent.py and scripts/process.py) that automate administrative tasks such as creating, launching, and monitoring access recertification campaigns via the Saviynt REST API.
[CREDENTIALS_UNSAFE]: Management scripts utilize command-line arguments to receive administrative usernames and passwords, which is a potential security risk as credentials may be visible in process listings or shell history.
[EXTERNAL_DOWNLOADS]: The skill relies on well-known Python packages including requests and pandas to facilitate communication with the Saviynt cloud infrastructure and process identity data.
[SAFE]: The scripts include a configurable option to bypass TLS certificate validation (SKIP_TLS_VERIFY), which is intended for lab environments but constitutes a security risk if enabled in production. No evidence of obfuscation, persistence, or data exfiltration was found.
[SAFE]: The skill processes external data from the Saviynt API (e.g., user entitlements and risk scores). While it lacks explicit boundary markers to prevent potential indirect prompt injection from malicious API metadata, the risk is considered low given the intended administrative use case and lack of automated execution of arbitrary content.

performing-access-recertification-with-saviynt