The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from external URLs to detect project flags and perform detailed quality and risk assessments.
Ingestion points: External content is fetched during Step 1 (via steps/01-flag-detection.md and a referenced URL fetch cascade) and is subsequently processed by three core agents and up to six conditional agents.
Boundary markers: The instructions do not specify the use of delimiters or clear boundaries to isolate external content from the agent's system instructions.
Capability inventory: The skill possesses file-system write capabilities (saving reports to OUTPUT_FOLDER) and executes a local CLI tool (aqe) for persistence.
Sanitization: There is no evidence of sanitization or escaping of the fetched URL content before it is passed to the sub-agents for analysis.
[COMMAND_EXECUTION]: The skill executes local commands and scripts as part of its operational workflow.
Evidence: Step 7 (steps/07-learning-persistence.md) invokes the aqe memory store command to persist findings to a local memory store. Additionally, the skill's changelog references an automated browser cascade executed via scripts/fetch-content.js.

qcsd-ideation-swarm