web-service-rn-python
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute a local verification script via the command
python3 scripts/workflow.py validateinSKILL.md. - [EXTERNAL_DOWNLOADS]: In
references/source-map.md, the skill references external code and documentation repositories. One such reference is to an untrusted source,nealepetrillo/claude-skills-fastapi, which the agent is instructed to access for framework guidance. - [INDIRECT_PROMPT_INJECTION]: The skill determines which instructions to load based on 'Task signals' from user input. This process lacks sanitization and boundary markers, creating a vulnerability where untrusted input could manipulate the agent's routing behavior.
- Ingestion points: User-provided task descriptions used in
SKILL.mdandreferences/current-guidance.mdto select branches. - Boundary markers: The instructions do not define delimiters or negative constraints to prevent the agent from following instructions embedded within the task signals.
- Capability inventory: The skill is capable of executing a local validation script and determining the active instruction set by loading other platform skills.
- Sanitization: The input used for routing decisions is not validated or sanitized.
Audit Metadata