memory-and-handoff
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by persisting distilled information from the workspace and agent interactions into durable session memory files. \n
- Ingestion Points: The skill reads from and writes to .cc10x/ memory files, as well as repository artifacts in docs/plans/ and docs/research/. \n
- Boundary Markers: Templates define specific markdown headings (e.g., ## Decisions, ## Learnings) and HTML comments to provide structure and separation for stored context. \n
- Capability Inventory: The skill is granted access to Read, Write, Edit, Bash, Grep, and Glob tools. \n
- Sanitization: Explicit rules mandate the redaction of credentials and PII from artifacts, utilizing a redacted:secret placeholder. \n- [COMMAND_EXECUTION]: The skill utilizes shell commands to manage its internal directory structure and export temporary handoff artifacts. \n
- Evidence: Instructions include the use of mkdir -p .cc10x for state management and the creation of handoff files in the system's temporary directory (/tmp/) using shell-generated filenames.
Audit Metadata