The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses dynamic context injection (!command syntax) to execute shell commands like cat and git upon loading. These are used to populate the agent's context with sprint logs and reviews from the local repository. Additionally, the skill provides a Bash script to automate git commits of the retrospective results.
[PROMPT_INJECTION]: An indirect prompt injection surface was identified where the skill processes data from external files.
Ingestion points: Reads content from docs/sprint-reviews/*.md and .claude/sprint-log.md in SKILL.md.
Boundary markers: None used; content from these files is injected directly into the agent's context without delimiters or warnings to ignore embedded instructions.
Capability inventory: The skill has access to Write, Edit, and Bash tools (specifically for cat, git, date, and grep) as defined in SKILL.md.
Sanitization: No validation or filtering is performed on the ingested file content before it is processed by the agent.

retro