cekura-infra-test-suite
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading the user's entire codebase to identify components and logic (e.g., Phase 1 and Phase 2). This content is subsequently used to generate test scenario descriptions, expected outcomes, and an orchestration script.
- Ingestion points: The agent reads the local codebase as defined in
phase1-explore.mdandphase2-analyze.md. - Boundary markers: The skill does not define explicit delimiters or 'ignore embedded instructions' warnings for the codebase content it processes.
- Capability inventory: The agent can write to the local filesystem (
/tmp/intermediate files andinfra_test_run.sh) and invokemcp__cekura__*tools to create resources and execute tests on the Cekura platform. - Sanitization: The instructions do not specify sanitization, escaping, or validation of the codebase content before it is interpolated into prompts or generated scripts.
- [DYNAMIC_EXECUTION]: The skill generates a shell script (
infra_test_run.sh) or Python script (infra_test_run.py) in Phase 5 to orchestrate the test suite. This script is constructed by concatenating scenario IDs, user-provided deployment commands, and template logic. While this is the primary purpose of the skill, generating executable code from interpreted data presents a standard vulnerability surface.
Audit Metadata