The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data which creates a surface for indirect prompt injection. It aggregates campaign details from upstream skills and incorporates user-provided notes into final readiness reports.
Ingestion points: Data enters from upstream skills including scope-audit, fte-capacity-sizing, and margin-simulation, as well as user-provided checklist 'notes' (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions to the agent to disregard embedded instructions within the processed data.
Capability inventory: The skill executes a local Python script (scripts/qa_scorer.py) and has the capability to trigger downstream skills like executive-growth-memo.
Sanitization: No sanitization or validation of the external content is performed before interpolation into the final visual and markdown outputs.
[COMMAND_EXECUTION]: The skill involves the execution of a local Python script to perform its core functionality of campaign scoring.
Evidence: The SKILL.md file provides instructions and examples for executing scripts/qa_scorer.py using the system Python interpreter.

campaign-launch-qa