The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands using the openspec CLI to create changes, monitor status, and fetch artifact instructions. User-supplied descriptions are used to derive command arguments, which may pose a risk if the agent fails to correctly sanitize or format the input.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes and follows instructions retrieved dynamically from the openspec instructions command output.
Ingestion points: Data is ingested from the output of the openspec status and openspec instructions commands, as well as from reading existing artifact files (SKILL.md).
Boundary markers: The skill does not use explicit boundary markers or "ignore instructions" wrappers when processing data from the CLI or local files.
Capability inventory: The skill possesses the capability to execute shell commands (bash) and write to the local file system at paths determined by the CLI (SKILL.md).
Sanitization: There is no evidence of strict validation, escaping, or filtering of the content retrieved from external tool outputs before it is used to guide the agent's actions.

openspec-propose