The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the openspec command-line interface via a restricted Bash environment (openspec:*). This scoping follows security best practices by limiting the agent's shell access strictly to the required toolset.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface by processing data from an external source.
Ingestion points: Data is ingested from the output of openspec instructions (including context, rules, and templates) within SKILL.md (Step 4a).
Boundary markers: Explicit boundary markers or warnings to the agent to ignore instructions within the ingested data are absent.
Capability inventory: The skill has the ability to execute openspec CLI commands and write files to the local file system at resolved paths.
Sanitization: No specific sanitization or validation of the ingested instructions is performed before use.
[SAFE]: The skill's behavior is consistent with its documentation. It does not attempt to access sensitive system files, hardcode credentials, or perform unauthorized network operations. The use of the openspec tool is restricted and appropriate for the skill's intended purpose.

openspec-propose