The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to analyze and review user-provided activity source code for idempotency hazards. This creates an attack surface where malicious instructions could be embedded in the code being reviewed.
Ingestion points: The agent ingests untrusted source code for Dapr Workflow activities via user prompts or file reads as described in the 'Rules' section.
Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore embedded commands within the code being analyzed.
Capability inventory: The skill explicitly instructs the agent to use 'Catalyst MCP tools' to perform its tasks, which may include reading run history or interacting with project infrastructure.
Sanitization: There are no specific sanitization or escaping instructions provided for handling the external code content before interpolation into the agent's reasoning process.
[SAFE]: The skill references vendor-specific infrastructure and tools which are consistent with the author's identity ('diagridio').
The link to 'https://catalyst.diagrid.io' is a legitimate vendor resource for the Diagrid Catalyst platform.
The reference to the 'dapr-ext-workflow' Python package is consistent with the skill's purpose of supporting Dapr-based development.

catalyst-activity-idempotency