The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external data (skill definitions and registry metadata) that could potentially contain indirect prompt injections targeting the catalog curation process.
Ingestion points: The agent is instructed to read skills-registry.json and individual SKILL.md files located in the skills/ directory.
Boundary markers: Absent. There are no instructions to use delimiters or to ignore embedded instructions within the processed data.
Capability inventory: The skill has access to shell execution via make, npm, git, and various local scripts (sync-skills.mjs, install.sh).
Sanitization: Absent. The curation workflow does not include steps to sanitize or validate the content of the skills being curated before processing them.
[COMMAND_EXECUTION]: The skill involves the execution of multiple shell commands and local repository scripts to perform curation tasks.
Development tools: Instructs the use of make list, make setups-check, git diff --check, and npm run build.
Local scripts: Mentions the use of scripts/sync-skills.mjs, install.sh, and install.ps1 for synchronization and setup tasks.

skill-catalog-curation