The Agent Skills Directory

[SAFE]: The skill's instructions are consistent with its stated purpose and do not contain malicious patterns or requests for sensitive information.
[NO_CODE]: The skill consists solely of instructions and architectural diagrams in a markdown file; it does not include any scripts or executable files.
[EXTERNAL_DOWNLOADS]: The instructions guide the agent to retrieve external documentation from a2a-protocol.org and research SDKs within the a2aproject GitHub organization to support the implementation process.
[PROMPT_INJECTION]: The implementation patterns involve processing untrusted data from external sources, which introduces an indirect prompt injection surface.
Ingestion points: The agent is instructed to fetch and process agent-card.json and response payloads from other agents.
Boundary markers: The skill does not provide instructions on using delimiters to isolate untrusted agent output from system prompts.
Capability inventory: The skill is configured to allow tools such as Bash, Write, and WebFetch.
Sanitization: No specific data validation or sanitization protocols are mentioned for the external inputs received during agent discovery or task execution.

a2a-client