The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the adaptyv-sdk package directly from a remote GitHub repository (https://github.com/adaptyvbio/adaptyv-sdk.git) using the uv package manager.
[REMOTE_CODE_EXECUTION]: Installation and subsequent import of the adaptyv-sdk from a remote repository involves the execution of code provided by that source during both setup and runtime.
[INDIRECT_PROMPT_INJECTION]: The skill processes experimental data and target information retrieved from the Adaptyv Bio API, which introduces a potential attack surface for indirect prompt injection if the API responses were to contain maliciously crafted instructions.
Ingestion points: Data retrieved from endpoints such as GET /experiments/{experiment_id}/results and GET /targets (referenced in references/api-endpoints.md).
Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" wrappers for data returned from the API.
Capability inventory: The skill utilizes the adaptyv Python SDK and shell commands (curl, uv, export) to manage laboratory experiments as described in SKILL.md.
Sanitization: The skill includes a high-level instruction to "sanitize molecules before computing" under the Agent operating procedure in SKILL.md.

adaptyv