The Agent Skills Directory

[COMMAND_EXECUTION]: The skill attempts to execute a local Python script (resolve_customization.py) from the project's _bmad/scripts/ directory to resolve configuration merges. While this allows for flexible customization, it relies on the integrity of the files within the user's project root.
[EXTERNAL_DOWNLOADS]: The workflow references the potential invocation of external skills, specifically bmad-advanced-elicitation and bmad-party-mode, to assist with generating and reviewing implementation rules. These references occur during interactive steps of the workflow.
[DATA_EXFILTRATION]: The skill reads various project files (e.g., package.json, requirements.txt, architecture.md) to gather context. This information is used exclusively to generate a local project-context.md file and is not transmitted to external domains or third-party services.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection:
Ingestion points: Reads contents from {project-root}/**/project-context.md, {planning_artifacts}/architecture.md, and common package manifest files (package.json, requirements.txt, etc.).
Boundary markers: No explicit delimiters or instructions are used to separate untrusted file content from the agent's internal instructions during discovery.
Capability inventory: Capable of executing shell commands via python3 and performing file writes to the local filesystem.
Sanitization: No explicit sanitization or filtering of the ingested file content is documented before processing.

bmad-generate-project-context