The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to automate the discovery of existing project assets and to facilitate file delivery.
Evidence: In assets/modes/ANALYZE.md, shell commands including ls, sort, and head are used to retrieve the most recent analysis and architecture decision records from the output directory.
Evidence: The skill's delivery workflow in SKILL.md includes an option to move files to a custom destination provided by the user.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of untrusted project documentation.
Ingestion points: The skill reads project-level files such as AGENTS.md, CLAUDE.md, and README.md, along with previous analyses and ADRs located in the user's output directory.
Boundary markers: No specific delimiters or instructions to ignore instructions within the ingested content are present in the prompts.
Capability inventory: The skill is authorized to use Bash, Write, Read, Glob, and Grep, which could be exploited if malicious instructions are successfully injected.
Sanitization: There are no mechanisms described for sanitizing or validating the content of the markdown files before they are processed by the agent.

growth-architect