The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from various project files which serves as a potential surface for indirect instructions.\n
Ingestion points: The skill reads .specify/extensions.yml, spec.md, plan.md, tasks.md, constitution.md, openapi.yaml, CLAUDE.md, and source code files.\n
Boundary markers: No specific delimiters or warnings for the agent to ignore embedded instructions within these files are mentioned.\n
Capability inventory: The skill can append findings to tasks.md, execute the local .specify/scripts/bash/check-prerequisites.sh script, and trigger other slash commands via EXECUTE_COMMAND.\n
Sanitization: The skill does not explicitly describe sanitization or validation logic for the content read from files or the command IDs extracted from the extensions configuration.\n- [COMMAND_EXECUTION]: The skill executes a local prerequisite check script located at .specify/scripts/bash/check-prerequisites.sh. This is an expected operation for project-specific tooling provided by the vendor.\n- [DYNAMIC_EXECUTION]: The skill implements a mechanism to execute hooks defined in the .specify/extensions.yml file. It dynamically constructs command identifiers (e.g., mapping speckit.git.commit to /speckit-git-commit) and triggers them using the EXECUTE_COMMAND directive.

speckit-docs-converge