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 skills/temiroff/blacknode/blacknode-developmentnpx skills add temiroff/Blacknode --skill blacknode-developmentgit clone --depth 1 https://github.com/temiroff/BlacknodeWhat 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 | $0.00103 | $0.01985 |
| Opus 5 | $0.00051 | $0.00992 |
| Sonnet 5 | $0.00021 | $0.00397 |
| Haiku 4.5 | $0.00010 | $0.00198 |
Grade A, and why
blacknode-development 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 2d ago.
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 — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Blacknode Development
Develop the smallest reusable layer that owns the requested capability, protect saved-workflow compatibility, and prove the result through both focused tests and a representative workflow when applicable.
Start Here
- Read the nearest
AGENTS.md; package repositories have scoped guidance. - Inspect the current implementation, tests, and public documentation before choosing a layer.
- Check Git status in the exact worktree being changed.
packages/*are independent repositories, not tracked children of the core repository. - Decide whether the capability belongs in core, an extension package, a
user-local custom node, or a workflow-local
PythonFn. - For a hardware, transport, simulation, or replay provider, read
docs/provider-authoring.mdand inspect the live capability contract. - Implement the narrowest coherent change, add tests, and update public docs.
- Build a small workflow/template when it materially proves discovery, ports, runtime behavior, or package resolution.
Choose the Ownership Layer
| Need | Put it here |
|---|---|
| Graph types, execution, validation, registry, CLI, MCP, exporters | python/blacknode/ |
| Browser/editor behavior or UI | editor/ |
| Editor HTTP API, runtime services, replay | editor-server/ |
| Portable Rust graph/runtime/CLI behavior | crates/ |
| Broad, dependency-light node | python/blacknode/nodes/ |
| Hardware, GPU, ROS, vision, or large optional dependency | extension package |
| User-owned local reusable behavior | custom-nodes/ |
| One graph's small adapter | PythonFn in the workflow |
Do not duplicate a domain implementation in core and a package. Keep the core contract generic and place transport- or hardware-specific implementations behind package nodes and runtime adapters.
Preserve Hardware Replaceability
- Keep base robot contracts, profiles, and capability inspection functional with no vendor driver, sensor SDK, ROS installation, simulator, or accelerator package installed. Missing providers return a structured unavailable state and do not prevent unrelated capabilities from loading.
- Make applications, skills, planners, and controllers consume stable capability contracts rather than vendor SDKs, device paths, transports, or component implementations.
- Bind capability providers through robot profiles and component configuration. A compatible camera, LiDAR, actuator bus, simulator, or compute-provider swap must preserve mission logic, workflow connections, and semantic node names.
- Bind calibration, limits, and sensor extrinsics to stable physical identity, not merely a component or driver type.
- Supply a mock or replay provider for every hardware capability. Do not claim providers are interchangeable until the same contract suite passes against the mock and every supported provider.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 202 lines · 103 tokens per session scan A d8615931dd60
blacknode-development is a skill published in the GitHub repository temiroff/Blacknode (23 stars, last pushed 2d ago), licensed Apache-2.0. It adds 103 tokens to every session and 1,985 once invoked, about $0.0005 per session on Opus 5. 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 skills, from other repositories
designer_guide
Guidance for creating and editing designs using the designer tool.
lynkr
Universal LLM gateway with intelligent routing, Graphify code intelligence, Distill compression, routing telemetry, Code Mode, and 12+ provider support. 60-80% cost reduction for Claude Code, Cursor, and Codex.
agents_guide
Guidance for delegating focused work to child Agents.
summarize-assessment
Use after the napkinmath pipeline has produced parameters/bounds/scenarios/montecarlo JSON to generate a plan assessment (assessment.md) — a thin interpretation layer over the intermediary artifacts. Emits a JSON manifest, a provenance map, gate verdicts (Critical / Fragile / Marginal / Robust), failure drivers…
data_analyst
Analyses datasets with professional rigour — statistical summaries, clear narratives, and well-chosen visualisations.
updater_guide
Guidance for checking for and installing Row-Bot updates.