The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute specific shell scripts located in the local project directory, such as ./.specify/scripts/bash/create-beads-epic.sh and ./.specify/scripts/bash/update-tasks-with-beads-ids.sh. These scripts handle the core synchronization logic between project files and the Beads database.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by ingesting and acting upon data from user-editable files like tasks.md and spec.md without specified sanitization.
Ingestion points: Reads content from specs/001-feature/tasks.md and specs/001-feature/.beads-epic-id to perform bulk task imports and ID linking.
Boundary markers: Absent. The skill does not define delimiters or specific warnings to ignore instructions embedded in the task descriptions or specs.
Capability inventory: The skill is permitted to use Bash for command execution, as well as Read and Edit for file system interactions.
Sanitization: Absent. There is no mention of validating, escaping, or filtering the data extracted from the external markdown files before it is passed to scripts or used in subsequent logic.

beads-integration