The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute launcher scripts including solo-research.sh and solo-dev.sh. It also provides instructions for monitoring via shell commands like tail -f, cat, and watch -n2 -c solo-pipeline-status.sh.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection or command injection because user-provided input from $ARGUMENTS is parsed and passed directly into shell execution contexts for the research and dev pipelines.
Ingestion points: User-supplied input via $ARGUMENTS in SKILL.md.
Boundary markers: None identified; input is passed directly to script arguments.
Capability inventory: The skill has access to Bash (command execution) and Write (filesystem modification).
Sanitization: No evidence of input validation, escaping, or sanitization before passing strings to the shell launcher scripts.
[DATA_EXFILTRATION]: The skill reads from and writes to state files located at .solo/pipelines/solo-pipeline-{project}.local.md and the global fallback ~/.solo/pipelines/solo-pipeline-{project}.local.md. While used for tracking pipeline progress, this grants the agent access to manage configuration data in the user's home directory.

solo-pipeline