The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to ingest and act upon data from the codebase's history and internal documentation, creating an attack surface for instructions embedded in data.
Ingestion points: The agent is instructed to analyze git log output in references/hotspot-audit.md and to read and prune repository documentation such as AGENTS.md and CLAUDE.md as described in references/doc-gardening.md.
Boundary markers: No explicit delimiters or instructions are provided to ensure the agent ignores or sanitizes malicious instructions that might be embedded in the repository content.
Capability inventory: The skill assumes the agent has the ability to execute shell commands (git log) and modify files throughout the repository to reorganize the codebase.
Sanitization: There is no requirement for sanitization or validation of the ingested repository data mentioned in the instructions.
[COMMAND_EXECUTION]: The skill provides a shell command pipeline for the agent to execute when auditing the repository for churn hotspots.
Evidence: git log --since="3 months ago" --name-only --pretty=format: | grep -v '^$' | sort | uniq -c | sort -rn | head -25 in references/hotspot-audit.md.

parallel-agents