The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as pnpm skill-tools lint <skill-name> and suggests the manual execution of node packages/skill-tools/dist/index.mjs eval. These commands use the provided skill name as an argument and are intended for project-local auditing and evaluation.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes untrusted data from other skills' SKILL.md and evals.json files.
Ingestion points: The workflow steps 1 and 6 in SKILL.md involve reading and auditing external skill files.
Boundary markers: The instructions do not specify the use of boundary markers or delimiters when reading these external files.
Capability inventory: The skill utilizes shell command execution via the pnpm skill-tools utility and node scripts located within the monorepo.
Sanitization: There is no explicit sanitization logic for the content read from external skills or for the $ARGUMENTS input before it is used in command strings.

skill-reviewer