The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read, search, and analyze external files from a user's codebase, including source code, tests, documentation, and configuration files. This data is then used to ground the agent's thinking and generate design artifacts. Maliciously crafted content within these external files could potentially influence the agent's reasoning or contaminate the output of generated artifacts.
Ingestion points: The agent is instructed to read relevant OpenSpec artifacts, source code, tests, documentation, and openspec/config.yaml files within the workspace.
Boundary markers: There are no explicit instructions to use delimiters or ignore instructions embedded within the ingested codebase files.
Capability inventory: The skill utilizes the openspec CLI to scaffold new changes, update proposals, and create design/specification artifacts.
Sanitization: The skill does not define specific sanitization or filtering procedures for the content read from the codebase, though it does instruct the agent not to reproduce private context directly.
[COMMAND_EXECUTION]: The skill relies on the openspec CLI tool to perform its primary functions. It includes detailed instructions for managing store states, listing changes, and generating instructions for artifact creation. While these are intended behaviors, they involve the execution of shell commands based on project state and user input.

openspec-explore