The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill executes shell commands at load time to gather environment metadata.
Evidence: The frontmatter uses the dynamic context injection syntax ! to run git branch --show-current, git rev-parse --short HEAD, and git status -sb | head -1 when the skill is loaded.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from external CI logs and automated fix generators, creating a surface for indirect instructions to influence agent behavior.
Ingestion points: The skill ingests suggestedFix, suggestedFixReasoning, and taskOutputSummary from responses generated by the ci-monitor-subagent which polls Nx Cloud.
Boundary markers: There are no explicit delimiters or warnings to the LLM to ignore instructions that may be embedded within the fix descriptions or task summaries.
Capability inventory: The skill can execute shell commands (nx run), apply remote patches (nx-cloud apply-locally), and perform write operations to the repository (git commit, git push).
Sanitization: No sanitization or validation of the external fix content is documented before it is applied or committed.
[DYNAMIC_EXECUTION]: The skill is designed to download and apply code patches from a remote service and execute build tasks dynamically based on CI results.
Evidence: The skill invokes nx-cloud apply-locally <shortLink> to apply patches from a remote identifier and runs dynamically determined tasks using nx run <taskId>.
[COMMAND_EXECUTION]: The skill performs various shell operations involving version control and package managers.
Evidence: Execution of git commit, git push, pnpm install, npm install, and yarn install are defined as standard workflows for handling CI failures.

monitor-ci