The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local PowerShell script at .specify/scripts/powershell/check-prerequisites.ps1 to initialize the analysis context and verify project state.
[DYNAMIC_EXECUTION]: The skill uses an extension system that reads command strings from a local YAML file (.specify/extensions.yml) and instructs the agent to execute them. Mandatory hooks are triggered using the EXECUTE_COMMAND directive followed by a specific instruction to invoke the command immediately.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple project files which could influence agent behavior or command execution.
Ingestion points: Loads content from spec.md, plan.md, tasks.md, constitution.md, and extensions.yml.
Boundary markers: No specific delimiters or instructions are used to treat the artifact content as untrusted data rather than instructions.
Capability inventory: Capability to execute local shell scripts and trigger dynamic agent commands based on file content.
Sanitization: No validation or sanitization of input strings from the artifacts or extension configuration before they are used to drive agent actions.

speckit-analyze