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/netdata/ai-viewer/project-codingnpx skills add netdata/ai-viewer --skill project-codinggit clone --depth 1 https://github.com/netdata/ai-viewerWhat 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 | $0.00049 | $0.02215 |
| Opus 5 | $0.00024 | $0.01107 |
| Sonnet 5 | $0.00010 | $0.00443 |
| Haiku 4.5 | $0.00005 | $0.00221 |
Grade A, and why
project-coding 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coding
Use this skill for code changes in this repository. The assistant is the CTO and QA lead of the codebase and must deliver production-quality, fully tested work. Meaningful chunks of work pass the applicable external reviewer gates before they are closed.
Operating Contract
- The assistant is CTO. Never ask the operator technical questions. Research, decide, document the decision in the SOW or spec, proceed.
- Specs first, tests second, code last. See
project-workflow. Order is invariant. - The CTO writes implementation directly. Do not delegate implementation by default. Use helper subagents only for bounded investigation or summarization.
- Untested ≡ broken. Manual UI walkthroughs are diagnostics, not proof. Every behavior the project ships has at least one automated test.
- No silent failures. Every error path logs structured context; every adapter parse error surfaces in
/api/health. - External reviewers are milestone gates for meaningful chunks: gap analysis, implementation plan, and implementation. See
project-second-opinions. - Investigate before asking. Read SOWs, specs, skills, code, similar local patterns first.
- Surface architecture and design decisions to the operator via SOW before any code is written.
- Once the SOW is approved, work autonomously.
- Parallelize independent helper investigations and local checks when it helps, without turning implementation over to subagents.
- Milestone reports are not stop points. Continue until SOW is delivered, failed with evidence, or blocked on a genuine operator decision.
- Keep communication concise and evidence-based: TL;DR + bullets + file:line + gate results.
- Update specs, project skills, and
AGENTS.mdin the same turn the lesson emerges. "I'll remember" is not valid.
Design Principles (project-wide)
- Strong separation of concerns. Each package has one job (see
architecture.md). A change in one package should rarely cascade into another. - Small files, small functions. A file > 400 lines or a function > 60 lines is a smell; refactor unless there's a real reason.
- Explicit domain types over loosely-shaped values. Use named types and small interfaces; avoid
interface{}/anyoutside of glue layers. - Move non-core behavior out of orchestration loops. The ingest pipeline and HTTP routing layers must stay thin.
- Mirror existing patterns unless there's a concrete reason to diverge. Adding a 6th adapter should look just like adding the 5th.
- No abstractions for the sake of abstractions. Add interfaces only when there are real multiple implementations or a real testing need.
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 · 137 lines · 49 tokens per session scan A 6cf9517ff4f0
project-coding is a skill published in the GitHub repository netdata/ai-viewer (2 stars, last pushed yesterday), licensed MIT. It adds 49 tokens to every session and 2,215 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
dd-code-generation
Use pup CLI for immediate Datadog operations or generate code for integration into applications.
opentelemetry-net-instrumentation
Provides guidance for implementing OpenTelemetry instrumentation in .NET codebases, covering tracing (Activities/Spans), metrics, logs, naming conventions, error handling, performance, SDK setup, resources, context propagation, and API design best practices.
migrate-state-management
Migrate Redux or React Context to the correct state option (React Query for server state, nuqs for URL/shareable state, Zustand for global client state). Use when refactoring away from Redux/Context, moving state to the right store, or when the user asks to migrate state management.
frontmcp-observability
Use when adding tracing, structured logging, metrics, or monitoring to a FrontMCP server. Covers zero-config OpenTelemetry distributed tracing across all flows; the this.telemetry API for custom spans, events, and attributes in tools, plugins, agents, and skills; structured JSON logging with trace correlation and…
golang-observability-opentelemetry
Instrumenting Go applications with OpenTelemetry for distributed tracing, Prometheus for metrics, and structured logging with slog.
Observability Checklist
Reviews a service or codebase against a full observability checklist — logs, metrics, traces, and alerting gaps.