The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including git diff and grep to analyze repository state. It also suggests the use of bin/rails db:migrate to resolve identified drift.
[PROMPT_INJECTION]: The skill exhibits an attack surface for Indirect Prompt Injection (Category 8) by ingesting and processing untrusted data from repository diffs.
Ingestion points: The skill reads file names and diff content from db/migrate/ and db/schema.rb using git diff.
Boundary markers: No specific delimiters or instructions are used to separate the ingested diff content from the agent's instructions, nor are there warnings to ignore instructions embedded in the data.
Capability inventory: The agent has the capability to run git commands and suggest/execute Rails migration commands, which could be exploited if a malicious migration file is processed.
Sanitization: There is no evidence of sanitization, validation, or escaping of the ingested diff data before it is processed by the model.

schema-drift-detector