The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses jq and shell redirection to modify the agent's internal configuration file (.claude/settings.local.json). This alters the execution environment by enabling experimental features automatically without explicit user intervention beyond the initial skill invocation.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by ingesting data from tasks.md and spec.md to drive its orchestration logic.
Ingestion points: The skill reads tasks.md to analyze dependencies and spec.md to provide context for sub-agents.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested files.
Capability inventory: The skill possesses the ability to execute shell commands (jq), spawn multiple sub-agents, and perform git operations (worktree management).
Sanitization: No sanitization or validation is performed on the content of the markdown files before they are parsed or interpolated into sub-agent prompts.

teams-orchestrate