The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill is instructed to run a shell script located at .specify/scripts/bash/setup-plan.sh from the repository root. This allows for the execution of arbitrary code contained within the project being analyzed.
[COMMAND_EXECUTION]: The skill implements an extension hook system that identifies command names from the .specify/extensions.yml file and automatically executes them using the EXECUTE_COMMAND: directive. This allows a project configuration file to trigger arbitrary command execution within the agent session.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted sources within the repository, including project configuration files and feature specifications, which could influence agent behavior.
Ingestion points: Reads and parses .specify/extensions.yml, FEATURE_SPEC, and .specify/memory/constitution.md.
Boundary markers: None identified; instructions do not provide delimiters or warnings to ignore instructions embedded in these files.
Capability inventory: The skill has the capability to write multiple Markdown files to the workspace, execute shell scripts, and invoke dynamic commands.
Sanitization: Only minimal sanitization is performed by replacing dots with hyphens in hook names before invocation.
[DYNAMIC_EXECUTION]: The skill dynamically generates command identifiers based on content found in the extensions.yml configuration file and proceeds to execute them.

speckit-plan