The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill provides automated instructions to search for and extract sensitive information from common configuration files such as application.yml, application.properties, and pom.xml. This includes database connection strings, datasource configurations, and credentials for middleware components like Redis, RabbitMQ, and Nacos.
Evidence: Shell command pattern grep -r "url:|driver-class-name:" --include="*.yml" --include="*.yaml" --include="*.properties" used to find database secrets.
Evidence: Instructions in SKILL.md to "Document key configurations: datasource, Redis, MQ, Nacos, ports".
Evidence: Reference file references/spring-patterns.md highlights patterns for extracting environment variables like ${DB_PASSWORD} and ${REDIS_PASSWORD}.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process untrusted data from a project's filesystem, including Java source code, XML/YAML configuration files, and Markdown documentation. This allows for an attack where malicious instructions embedded in project files could be executed or adopted by the agent during analysis.
Ingestion points: The skill reads all files matching *.java, pom.xml, build.gradle, *.yml, *.properties, and README* within the targeted directory.
Capability inventory: The agent uses shell utilities like grep, find, cut, and sort to extract and process data from these files.
Boundary markers: None. The skill does not include instructions to the agent to treat file content as data rather than instructions.
Sanitization: No sanitization or validation of the ingested content is performed before it is used to generate the final documentation report.

project-analyzer