The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git commands (git log --oneline -20, git diff) to detect project history and changes.
[COMMAND_EXECUTION]: During the validation step in SKILL.md, the agent is instructed to "Run project commands mentioned in docs to verify they work," which involves executing arbitrary shell commands extracted from project documentation.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes and acts upon instructions found in external data sources.
Ingestion points: Processes content from README.md, CLAUDE.md, and rule files located in the .claude/ directory.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the processed documentation are present.
Capability inventory: The skill possesses the capability to execute shell commands, scripts, and build steps as part of its validation workflow.
Sanitization: There is no evidence of command sanitization, filtering, or verification against a whitelist before execution.

update-project