The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's instructions and implementation playbook describe workflows that rely on executing local shell commands, including git, jira-cli, and slack-cli, to automate the gathering and reporting of work status.
[DATA_EXFILTRATION]: The skill is designed to collect data from private repositories, task management systems, and personal knowledge bases (Obsidian), which it then transmits to an external AI service and a communication platform (Slack). This behavior is aligned with the skill's primary stated purpose.
[PROMPT_INJECTION]: The skill processes untrusted input from external sources such as commit messages and ticket descriptions. Due to the lack of explicit delimiters or sanitization logic in the generation prompt, it is potentially vulnerable to indirect prompt injection.
Ingestion points: Commit history (git log), Jira tickets (jira issues list), and Obsidian notes (obsidian_get_recent_changes).
Boundary markers: Absent from the provided AI prompt templates used for summarization.
Capability inventory: The skill can execute shell commands, write files to the disk, and perform network operations via local CLI integrations.
Sanitization: No validation or instruction-filtering mechanisms are present to sanitize the ingested data before it is processed by the AI.

team-collaboration-standup-notes