valuing-information
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates decision-making by executing a local Node.js script (
scripts/calc.js) to process statistical inputs. This execution is confined to the skill's own logic for calculating VOI, study precision, and sample sizes. - [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from JSON files and command-line inputs, creating an indirect prompt injection attack surface.
- Ingestion points: Input is accepted from
study.jsonandarms.jsonfiles, as well as via CLI flags such as--successes,--trials,--p, and--rsein theSKILL.mdand reference files. - Boundary markers: The provided files do not include explicit instructions to use delimiters or ignore embedded instructions within the processed data.
- Capability inventory: The skill environment allows for shell command execution via
nodefor its bundled scripts. - Sanitization: There is no documented mechanism for sanitizing or validating the inputs before they are passed to the script execution.
Audit Metadata