roslynskills-tight

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute the 'roscli' tool or 'scripts/roscli' script via Bash or PowerShell to perform operations such as symbol renaming, navigation, and code diagnostics. This includes the use of a 'query.batch' command which dynamically dispatches multiple sub-commands from a JSON-formatted input string.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to the ingestion and processing of untrusted content from the repository.
  • Ingestion points: Source code, project files, and search results are read into the agent's context through commands like 'nav.find_symbol', 'ctx.search_text', and 'analyze' calls.
  • Boundary markers: The skill does not define explicit delimiters or 'ignore' instructions to prevent the agent from accidentally following instructions that might be embedded within the processed repository files.
  • Capability inventory: The agent has the capability to execute shell commands via 'Bash' and modify the filesystem using the 'Edit' and 'Write' tools based on the output of its analysis.
  • Sanitization: There is no evidence of sanitization or validation of repository content before it is interpolated into prompts or used to guide agent actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 07:47 PM