The Agent Skills Directory

[COMMAND_EXECUTION]: The helper script scripts/audit-openocta.py uses the subprocess module to run git commands (such as git remote get-url and git rev-parse) to verify the integrity and origin of the local repository.
[INDIRECT_PROMPT_INJECTION]: The skill manages integrations with webhooks and instant messaging channels (such as Feishu, Telegram, and WeCom) that process external data, creating a potential surface for indirect prompt injection. The instructions mitigate this risk by requiring token authentication and advising the agent to treat incoming data as untrusted. Ingestion points: Webhooks (/hooks/agent) and IM channels. Boundary markers: Instructions to separate data from instructions and use token-based authentication. Capability inventory: Access to Bash, file system writes, and OpenOcta gateway operations. Sanitization: Documentation recommends validating source networks and payload types before processing.
[EXTERNAL_DOWNLOADS]: The skill fetches release metadata from GitHub's official API to assist users in selecting and verifying software assets. This operation targets a well-known and trusted service.

openocta