The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx to download and execute the @lanbaoshen/mcp-jenkins package from the NPM registry at runtime.
[COMMAND_EXECUTION]: The skill provides numerous patterns for executing arbitrary shell commands within Jenkins pipelines (sh), running Java applications (java -jar), and interacting with cluster management tools (kubectl).
[DATA_EXFILTRATION]: Documentation includes patterns for handling sensitive data, such as writing Kubernetes configuration to a temporary file (/tmp/kubeconfig) and managing Jenkins credentials via API, which presents a risk of exposure if the environment is not secured.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection.
Ingestion points: The jenkins_get_console_output tool in SKILL.md reads raw logs from external build processes.
Boundary markers: None identified; untrusted log content is processed without clear delimiters.
Capability inventory: The skill can trigger builds (jenkins_trigger_build), create jobs (jenkins_create_job), and execute shell scripts across various files.
Sanitization: No evidence of sanitization or filtering of build logs or job parameters before they are processed by the agent.

jenkins-ci