Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/rocketride-org/rocketride-server/rocketride_component_referencegit clone --depth 1 https://github.com/rocketride-org/rocketride-serverWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.06529 |
| Opus 5 | $0.00000 | $0.03265 |
| Sonnet 5 | $0.00000 | $0.01306 |
| Haiku 4.5 | $0.00000 | $0.00653 |
Grade A, and why
ROCKETRIDE_COMPONENT_REFERENCE scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 700 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RocketRide Component Reference
Last Updated: June 2026
How Component Information Is Organized
When the RocketRide VS Code extension is installed, it populates a .rocketride/ directory in your workspace:
.rocketride/
├── services-catalog.json # Summary of ALL components (name, class, lanes, descriptions)
├── schema/ # Detailed config schema for each component
│ ├── webhook.json
│ ├── llm_openai.json
│ ├── qdrant.json
│ └── ... # One file per component
└── docs/ # Documentation files (these files)
IMPORTANT: Always read .rocketride/services-catalog.json for the current list of available components. The catalog is the single source of truth: it is generated from the connected server and may contain components not listed in this document.
See also:
docs/README-nodes.mdis the node catalog grouped by category, with each node's lanes and the wire-vs-bind rule.docs/README-node-schema.mdexplains theservices.jsonmodel (lanes,preconfig/profiles,fields,shape) behind this reference and the canvas UI.
Reading the Catalog
Each entry in services-catalog.json has this structure:
{
"name": "component_provider_name",
"classType": ["category"],
"description": "What the component does",
"lanes": {
"input_lane": ["output_lane_1", "output_lane_2"]
},
"invoke": {
"llm": { "description": "LLM used by the component", "min": 1 },
"tool": { "description": "Tools available to the component", "min": 0 },
"memory": { "description": "Memory store", "min": 0, "max": 1 }
}
}
Key fields:
name: theprovidervalue you use in pipeline filesclassType: component category: source, data, text, image, audio, video, embedding, llm, store, database, tool, agent, memory, infrastructure, target, preprocessorlanes: the definitive reference for data flow. Each key is an input lane; its value array lists the output lanes produced. An empty array[]means the component consumes data with no lane output (e.g., storage, response). Source components use_sourceas the input lane keyinvoke: (optional) defines what control-plane connections the component requires or accepts. Each key is aclassType(e.g.,llm,tool,memory) withmin/maxconstraints and a description. Components withinvokeneed corresponding entries in thecontrolarray of the pipeline definition
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 700 lines · 0 tokens per session scan A 4e7867764b68
ROCKETRIDE_COMPONENT_REFERENCE is an agent published in the GitHub repository rocketride-org/rocketride-server (7,496 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 6,529 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
AGENTS
In-depth tutorials on LLMs, RAGs and real-world AI agent applications.
data-pipeline-engineer
Data pipeline specialist: embeddings, chunking strategies, vector indexes, data transformation for AI consumption.
datahub-guide
Phase 9 of the DataHub migration exposes the terminal's in-process pub/sub layer directly to LLM tool callers through four generic MCP tools. Everything streaming into any active widget — quotes, order books, news, vessel tracks, broker ticks, geopolitical events, agent outputs, LLM token streams — is observable from…
pixel-art-animation-reviewer
Independent reviewer of pixel-art ANIMATION quality (loop seamlessness, motion physics, multi-component motion, frame timing, period selection, particle determinism). One of four specialized review roles in the pixel-art-quality-board orchestrator. Use when the user asks to "check animation timing", "verify loop…
gke-cluster-runner
Launch a single TPU training workload on a GKE cluster via XPK, poll until completion or hang, capture xprof + HLO dumps to GCS, and report structured verdict signals back to the master agent. Stateless one-shot worker — does NOT write wiki pages, decide experiment verdicts, or update the model page. Use for every…
process-auditor
Delta-audits an autoresearch run (kernel family OR model lane) since the last audit and returns corrective findings to the runner's context. Runs as a self-rescheduling watcher armed ONCE at launch (start-experiment Step 9·0) — never re-dispatched per iteration by the runner. Read-only over pages, branches, receipts…