Manages Cargo.lock file updates and resolves --locked flag issues in CI/CD. Triggers when user mentions: - "cargo test --locked failed" - "cannot update the lock file" - "Cargo.lock is out of date" - "PR failed with --locked error" - "fix Cargo.lock"
Maintain the local OpenCode mirror for self-reference
Publish the openwork-orchestrator npm package with clean git hygiene. Triggers when user mentions: - "openwork-orchestrator npm publish" - "publish openwork-orchestrator" - "bump openwork-orchestrator"
Guide users through the get started setup and Chrome DevTools demo.
Security auditing for code, configs, and infrastructure. Use when the user wants to audit or improve security: scan for vulnerabilities (SQL injection, XSS, command injection, path traversal), detect hardcoded secrets and credentials, review auth and authorization, check dependencies for known CVEs, audit config files for insecure defaults, or generate security reports. Trigger on "security audit", "vulnerability scan", "code review for security", "find secrets", "check for vulnerabilities", "OWASP", "CVE", or questions about code security.
Start the OpenWork dev stack via Docker and verify real user flows via Chrome MCP. Triggers when user mentions: - "dev-up.sh" - "docker dev stack" - "verify in chrome mcp" - "test the real flow"
Guide users through browser automation setup using Chrome DevTools MCP as the primary path and the OpenCode browser extension as a fallback. Use when the user asks to set up browser automation, Chrome DevTools MCP, browser MCP, browser extension, or runs the browser-setup command.
Bridge between OpenWork UI and OpenCode runtime
Debug OpenWork sidecars, config, and audit trail
Internal helper contract for calling the codex-companion runtime from Claude Code
Internal guidance for presenting Codex helper output back to the user
Internal guidance for composing Codex and GPT-5.4 prompts for coding, review, diagnosis, and research tasks inside the Codex Claude Code plugin
Use when you need to visually interact with a GUI — test buttons, fill forms, verify visual layouts, fuzz web pages, automate user flows, take screenshots, or perform end-to-end QA on any application. Works on cloud VMs, Docker containers, local machines, and sandboxes. Install: pip install cua.
Compile TensorRT-LLM on a SLURM cluster. Covers submitting a batch job with a container image, monitoring the job, and verifying the build. Use when the user wants to compile TRT-LLM remotely via SLURM rather than on a local compute node.
ONLY for OpenAI Triton (@triton.jit) kernel development. NEVER use for CUDA C++ kernels, TileIR, or profiling tools (ncu, nsys). The user's request must involve Triton explicitly. Covers Triton-specific patterns: fused elementwise, reductions (softmax, LayerNorm, RMSNorm), tiled GEMM with triton.autotune, and flash attention. Workflow: design, write, verify (with fast-path for explicit requests).
Performance optimization coordination playbook. Contains specialist routing table, TileIR two-step pipeline, kernel generation specialist selection, prioritization criteria, and safe modification workflow. Use when the user asks to apply optimizations, write kernels, or improve performance. Covers both user-specified optimization and autopilot-driven iterative optimization.
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup, per-iteration statistics, and anti-pattern avoidance. Also covers NVTX annotation for labeling profiler timelines. NOT for: running or analyzing profiler tools (nsys, ncu, Nsight Systems, Nsight Compute), writing kernels (Triton, CuTe, CUDA), applying optimizations (CUDA Graphs, gradient checkpointing, fusion), or interpreting roofline/SOL% metrics. Triggers: "measure throughput", "benchmark this function", "time my training loop", "samples per second", "NVTX annotate", "instrument my dataloader", "data load time", "kernel timing", "how do I time".
Compile TensorRT-LLM on a compute node inside a Docker container. Use this when already on a compute node with GPUs visible.
Retrieve and diagnose CI test failures from TensorRT-LLM pull requests using the GitHub API and Jenkins testReport API. Use when the user asks about CI failures on a PR, wants to see failed test details, or needs stdout/stderr from a CI run.
