The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill identifies a JavaScript file at .claude/workflows/qcsd-development-review.js as its primary execution model. This script is dynamically loaded and executed by the platform's workflow harness to implement the core review pipeline.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a significant attack surface for indirect prompt injection as it ingests and processes arbitrary source code through multiple LLM-based specialist agents.
Ingestion points: Source code and test files are ingested from user-provided paths (SOURCE_PATH, TEST_PATH).
Boundary markers: No specific delimiters or instructions to ignore embedded malicious prompts are included in the agent directives.
Capability inventory: The swarm has the capability to execute shell commands (via aqe), spawn sub-agents, and perform extensive filesystem write operations for report generation.
Sanitization: There is no evidence of sanitization or filtering of the source code content before it is passed to the specialized analysis agents.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using a custom CLI utility called aqe (Agentic QE) for memory operations (aqe memory search and aqe memory store). Furthermore, the skill's evaluation and validation logic depends on the availability of the jq utility on the host system.

qcsd-development-swarm