axiom-run-tests
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill resolves a local helper binary named
axbuildby searching relative paths (e.g.,tools/axbuild) or the system PATH and executes it with shell arguments. - [DYNAMIC_EXECUTION]: The path to the
axbuildexecutable is determined at runtime based on the environment and loaded package structure, which is then used to spawn subprocesses. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external data including compiler logs, test results from
.xcresultbundles, and failure screenshots. - Ingestion points: Processes output from
xcodebuild,xcresulttool, and visual data from failure screenshots. - Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard potential instructions embedded within test logs or failure messages.
- Capability inventory: The skill has the ability to execute shell commands (
xcodebuild,xcrun), read/write files in temporary directories, and export attachments. - Sanitization: Uses
jqto safely parse structured JSON output from system tools, though the semantic analysis of logs remains an open surface.
Audit Metadata