The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from external URLs to generate summaries that are then submitted to a GitHub repository.
Ingestion points: External URLs provided by the user in SKILL.md.
Boundary markers: No specific boundary markers are used to isolate content fetched from the URLs.
Capability inventory: The skill utilizes gh issue create for network communication and performs file-write operations to /tmp via its shell scripts.
Sanitization: The scripts/submit_bulk.sh and scripts/submit_single.sh scripts perform validation on metadata fields, including length checks and the rejection of control characters (pipes, tabs) to prevent downstream parsing errors.
[COMMAND_EXECUTION]: The skill executes local shell scripts included in its package (scripts/submit_single.sh and scripts/submit_bulk.sh) to interact with the GitHub CLI (gh). This activity is restricted to the skill's primary purpose and involves vendor-owned repositories.
[SAFE]: The skill follows security best practices by requiring explicit user confirmation before performing submissions, checking authentication status before execution, and providing a dry-run mode for auditing actions.

honeycombo-submit