The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the openspec CLI via Bash to manage project changes and generate planning artifacts. Key commands include openspec new change, openspec status, openspec instructions, and openspec context. These operations are consistent with the skill's stated purpose of providing a CLI-driven workflow.
[INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting data from multiple untrusted or external sources to generate markdown artifacts (proposal, design, specs, and tasks).
Ingestion points: The process starts with user-provided descriptions of desired changes in Step 1. It also consumes JSON data (context, rules, and instructions) returned by the openspec instructions command in Step 5.
Boundary markers: The skill utilizes structured templates for file creation and includes explicit instructions for the agent to avoid copying metadata or constraint blocks (<context>, <rules>) into the final output files.
Capability inventory: The skill possesses the ability to execute shell commands and write content to filesystem paths determined dynamically by the CLI (resolvedOutputPath).
Sanitization: There is no evidence of explicit validation or sanitization of the content before it is written to the artifacts; the skill relies on the agent's instructions to maintain the planning boundary and follow the provided templates.

openspec-propose