performance-testing

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external JMeter (.jmx) and Gatling source files as input for conversion. This creates an attack surface where malicious instructions embedded in a test file could attempt to influence the agent's behavior during the conversion process.
  • Ingestion points: External JMeter test plans (.jmx) and Gatling simulation files (Java, Kotlin, Scala, JS, TS).
  • Boundary markers: The instructions do not specify explicit delimiters or boundary markers to isolate untrusted input content from the agent's instructions.
  • Capability inventory: The skill uses the postman CLI for collection runs, performance runs, and workspace operations (push, pull, migrate). It also performs file system write operations to generate the converted project artifacts (collection.json, environment.json, run.sh, MIGRATION.md).
  • Sanitization: The instructions do not define sanitization or filtering procedures for the content of the input test files.
  • [DYNAMIC_EXECUTION]: The skill translates logic from JMeter (Groovy/BeanShell) or Gatling simulations into Postman JavaScript snippets using predefined templates.
  • Evidence: references/jmeter-import/script-translation.md and references/gatling-import/mapping.md provide detailed instructions and templates for the agent to generate functional test scripts during migration.
  • [EXTERNAL_DOWNLOADS]: The skill describes scenarios where remote data files may be retrieved for use in the converted project.
  • Evidence: In references/gatling-import/feeders.md, the skill notes that jsonUrl(url) feeder sources should be downloaded (subject to user approval) to be incorporated into the local data directory.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands using the Postman CLI and Node.js for validation and testing purposes.
  • Evidence: references/jmeter-import/README.md and references/gatling-import/README.md instruct the agent to use node --check for script syntax validation and postman collection run for smoke testing the converted collection. The skill includes a specific requirement to obtain user confirmation before sending network traffic to non-local targets.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 08:29 PM