The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection (Category 8) as it processes data from external requirements sources.
Ingestion points: Data is ingested from Jira ticket summaries and descriptions via the brianna skill and from user-provided task descriptions in SKILL.md.
Boundary markers: Absent; the skill instructions do not define explicit delimiters or 'ignore' instructions when passing retrieved data to the implementation or planning phases.
Capability inventory: The skill has the capability to execute shell commands (git, gh, poetry, make), search the filesystem (Grep, Glob), and read files (Read).
Sanitization: Absent; there is no mention of sanitizing or validating the content retrieved from Jira or user descriptions before using them in command construction or architectural analysis.
[COMMAND_EXECUTION]: The skill directs the agent to execute various shell commands to manage the development lifecycle.
Evidence: Instructions include running git branch, git checkout, git pull, git push, gh pr create, poetry run pytest, and make ruff across SKILL.md and referenced guides.
Context: These commands are fundamental to the skill's purpose as a technical lead and are used to manage code changes and verify implementation.
[EXTERNAL_DOWNLOADS]: The skill retrieves data from an external Jira service.
Evidence: It uses the brianna skill to fetch ticket details (summary and description fields) as described in Phase 1 of the implementation workflow.
Context: This is documented neutrally as a requirement-gathering step and is limited to non-sensitive fields.

tech-lead