The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch additional documentation and predicate references from the vendor's official domain.
Evidence: curl https://docs.reticle.sh/predicates.md and curl https://docs.reticle.sh/llms.txt in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with the output of a running application (console logs, network responses, and signals), which represents an attack surface for instructions embedded in external data.
Ingestion points: The reticle_act_and_wait and reticle_assert tools ingest console output, network status/headers, and application signals into the agent's context (found in SKILL.md).
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the provided predicates.
Capability inventory: The skill utilizes reticle_act_and_wait for browser/application interaction and reticle_run for recording and saving flows (found in SKILL.md).
Sanitization: No sanitization or filtering of application-generated content is described.

agentic-tdd