meisijiya-phase-checkpoint
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from the
.omo/plans/<slug>.mdfile, creating an indirect prompt injection surface. It mitigates this by instructing the agent to treat the plan content as data only and to ignore any embedded directives or bypass attempts. Evidence Chain: 1. Ingestion points: The file.omo/plans/<slug>.mdis read to extract phase information. 2. Boundary markers: The skill includes explicit instructions to ignore 'hostile' text and treat content as data, not instructions. 3. Capability inventory: The agent has access to Bash, Write, and Read tools. 4. Sanitization: A regular expression check (^[0-9]+(\.[0-9]+)?$) is performed on the phase number extracted from the data before it is interpolated into shell commands. - [COMMAND_EXECUTION]: The skill uses the Bash tool to execute shell commands like awk and grep for data parsing. It includes a mandatory validation step where the phase identifier must match a strict numeric regex before being used in these commands, which prevents command injection vulnerabilities.
Audit Metadata