The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @t2000/cli package via npm and the addition of skills from the mission69b/t2000-skills repository.
[COMMAND_EXECUTION]: The skill relies on the t2 command-line tool to perform operations including wallet connection, job board retrieval, and file delivery (t2 job deliver). These commands interact with the local file system and remote marketplace servers.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from an external marketplace, creating a potential injection surface.
Ingestion points: The agent fetches and reads external data through t2 job board and t2 job spec <jobId> as described in SKILL.md.
Boundary markers: There are no explicit delimiters or warnings provided in the instructions to prevent the agent from obeying instructions embedded in the job specifications.
Capability inventory: The agent has the ability to execute shell commands using the t2 CLI and read/upload local file contents (out.md) for delivery.
Sanitization: No sanitization or validation of the job specifications is performed before the agent processes them.

t2000-earn