check-bin-obj-clash
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill instructs the agent to use the
dotnetCLI to query project properties and process binary logs. While this involves executing shell commands, it is the standard and expected method for analyzing MSBuild environments. - Indirect Prompt Injection Surface: The skill processes external
.binlogand text log files generated during build processes. Analyzing data from external sources presents a potential consideration where untrusted build output could attempt to influence the agent's context. The skill's instructions mitigate this by recommending the use of structured MCP tools for inspection rather than direct text processing where possible. - Ingestion points: External
.binlogandfull.logfiles (SKILL.md). - Boundary markers: Not explicitly defined for raw log parsing.
- Capability inventory: Uses
dotnetCLI for property queries and log replaying (SKILL.md). - Sanitization: The skill relies on structured MCP tools to interpret binary logs safely.
Audit Metadata