The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use local shell commands (git log, date) to retrieve project history and current timestamps. These operations are performed locally and are necessary for the skill's primary function of generating reports.
[INDIRECT_PROMPT_INJECTION]: The skill ingests git commit messages which are potentially untrusted. While an attacker could embed instructions in a commit message, the risk is negligible as the agent is instructed only to summarize the text into a static markdown file, not to interpret or execute it.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads local git logs to extract work summaries. This is the intended behavior for a reporting tool and does not involve accessing sensitive system credentials or exfiltrating data to external servers.

report-weekly