The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses grep and bash commands to perform discovery within the project's filesystem, scanning for Redis, Memcached, and database query patterns.
[CREDENTIALS_UNSAFE]: The discovery step specifically instructs the agent to grep for strings like REDIS_URL and MEMCACHE within environment files (e.g., .env, .env*) and configuration files. These variables often contain embedded passwords or tokens which would be exposed to the agent's context during discovery.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to its automated scanning and processing of untrusted local data.
Ingestion points: The skill reads various file types (.js, .py, .rb, .json, .yml, .env) throughout the repository using grep, Glob, and Read tools.
Boundary markers: The instructions do not define boundary markers or delimiters for the data being read.
Capability inventory: The agent has access to Bash, Write, and Edit tools, allowing it to execute commands or modify files based on instructions interpreted from the scanned data.
Sanitization: No sanitization or validation of the ingested code/configuration content is mentioned.

cache-strategy