The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the Bash and Task tools to execute system commands, create session directories in /tmp, and launch parallel agent tasks. While these are intended for orchestration, they represent a significant capability tier.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting results from codebase analysis (via claudemem) and passing them to subsequent agents.
Ingestion points: Code analysis results are written to and read from files like structure-map.md, test-gaps.md, and dead-code.md in SKILL.md.
Boundary markers: The skill does not explicitly define delimiters or instructions to ignore malicious content embedded within the analyzed code output.
Capability inventory: The skill has access to Bash, Task (parallel agent execution), and Write capabilities across its orchestration patterns.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the codebase before it is interpolated into task prompts.

claudemem-orchestration