runlens
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the
runlensbinary using Homebrew (itamaker/tap/runlens) or by downloading a release from the author's GitHub repository (github.com/itamaker/runlens-skill). These resources are associated with the skill author itamaker. - [COMMAND_EXECUTION]: The skill's primary function involves executing the
runlensbinary through the command line to process trace files. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes external JSONL trace files, creating an attack surface where untrusted data enters the agent context.
- Ingestion points: Untrusted data enters via the
-inputflag pointing to a JSONL file, as described inSKILL.mdandREFERENCE.md. - Boundary markers: Absent. There are no instructions for the agent to use delimiters or ignore embedded instructions when reading the tool's output.
- Capability inventory: The agent uses shell commands to run the
runlensbinary as described inSKILL.md. - Sanitization: Absent. The tool aggregates error messages and tool names for clustering and reporting but does not filter for malicious instructions before the agent processes the resulting output.
Audit Metadata