The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and audit repository documentation, which presents a surface for indirect prompt injection if an attacker can commit malicious instructions into the markdown files or code comments.
Ingestion points: The skill reads docs/AGENTS.md, implemented/ Agent Notes, and all .md files in the repository via git ls-files.
Boundary markers: There are no explicit instructions or delimiters used to separate the content being audited from the agent's core instructions.
Capability inventory: The skill has the capability to read/write/move files in the repository and execute shell commands (git, wc, sort, head).
Sanitization: No sanitization or validation of the processed documentation content is mentioned.
[COMMAND_EXECUTION]: The skill instructs the agent to execute specific shell command sequences for repository auditing.
Evidence: git ls-files '*.md' ':(exclude)vendor/**' | xargs wc -w | sort -rn | head -30 and git diff --check are explicitly mentioned as audit steps.

doc-standards