The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from untrusted external sources, creating a risk for indirect prompt injection.
Ingestion points: The skill pulls data from Git commit messages, Jira ticket titles/comments, and Obsidian vault contents (Daily Notes/Standup/2025-10-11.md).
Boundary markers: There are no explicit delimiters or instructions to the model to ignore potential commands embedded within the ingested commit history or Jira data.
Capability inventory: The skill instructions include the ability to execute shell commands (git, jira CLI, slack-cli) and interact with MCP tools for Atlassian and Obsidian.
Sanitization: No sanitization or validation of the ingested external content is mentioned before it is passed to the AI for summarization.
[COMMAND_EXECUTION]: The skill documentation and example scripts include the execution of shell commands to gather developer context.
Evidence: The 'Data Source Orchestration' and 'AI-Assisted Note Generation' sections define the use of git log, jira issues list, and potentially other CLI tools like slack-cli and gcal to automate the workflow.

dx-optimizer