cy-workflow-memory
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill maintains persistent workflow and task-level memory files which are treated as mandatory context for the run. This creates an attack surface where instructions embedded in processed data could influence the agent's behavior in subsequent executions.
- Ingestion points: SKILL.md dictates reading the shared workflow memory and current task memory files at the start of each run.
- Boundary markers: Absent. There are no instructions to use delimiters or ignore embedded instructions when processing the content of these memory files.
- Capability inventory: The skill involves reading and writing to local repository files. While the skill itself does not invoke shell commands, it provides persistent instructions that the agent is told to treat as mandatory context.
- Sanitization: Absent. The skill lacks validation or sanitization of the data written to or promoted between memory files.
Audit Metadata