The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the rg (ripgrep) command-line tool to search through the .agents/log directory for relevant design decisions based on user-provided keywords or file paths.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it reads existing log entries and treats those marked as done as binding constraints for future agent behavior. * Ingestion points: The agent reads .agents/log/index.md and individual entry files (e.g., YYYY-MM-DD-slug.md) from the .agents/log/ folder to inform its planning and implementation phases. * Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore potentially malicious instructions embedded within the design log files. * Capability inventory: The skill has the ability to read from and write to the local file system and execute shell commands via rg. * Sanitization: There is no evidence of sanitization, filtering, or validation of the content within the log entries before the agent incorporates them into its operational context.

design-log