The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the signadot CLI to perform operations such as creating plans, running executions, and fetching logs. These commands are limited to the specific functionality of the Signadot platform and align with the skill's primary purpose.
[CREDENTIALS_SAFE]: The instructions include specific guidance on handling sensitive information, directing the agent to use the --param-secret flag and avoid hardcoding secrets in YAML specifications or plain-text summaries.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading action bodies and execution logs from the Signadot platform. While this exposes a surface for indirect prompt injection, the skill's structured workflow and focus on technical authoring tasks mitigate the risk of the agent following malicious instructions embedded in tool outputs.

signadot-plan