The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for fetching runner scripts (quantum-loop.sh and quantum-loop.ps1) from the author's GitHub repository at https://raw.githubusercontent.com/andyzengmath/quantum-loop/main/templates/ to facilitate the execution phase of the pipeline.
[COMMAND_EXECUTION]: The instructions guide the user to grant execution permissions to the downloaded shell scripts using chmod +x and subsequently run them to manage the autonomous development loop.
[DATA_EXFILTRATION]: The skill performs read operations on local project configuration files, such as package.json, pyproject.toml, and go.mod, to extract metadata, technology stack information, and determine appropriate file paths for task generation.
[PROMPT_INJECTION]: The skill processes external Product Requirements Documents (tasks/prd-*.md) to extract user stories and requirements. This creates an indirect prompt injection surface as the agent lacks explicit boundary markers or sanitization logic for content ingested from these files.
Ingestion points: Reads requirements from tasks/prd-*.md (documented in SKILL.md).
Boundary markers: Absent; the agent is instructed to read the entire PRD directly into context.
Capability inventory: Generates task lists and verification commands within quantum.json (documented in SKILL.md).
Sanitization: Absent; the skill relies on the agent to parse and interpret the PRD content without filtering.

ql-plan