pinescript-plugin

Pine Script v6 for coding agents — backed by a real validator, not prose.

---

The problem

Ask any coding agent for a TradingView indicator and you get code that looks right, reads fluently, and does not compile.

// Confident. Fluent. Four compile errors.
l = line.new(x1=1, y1=2, x2=3, y2=4, colour=color.red)   // it's `color`, not `colour`
plotshape(cond, shape=shape.triangleup)                   // it's `style=`, not `shape=`
v = math.clamp(x, 0, 1)                                   // math.clamp doesn't exist in Pine
b = box.new(left=1, top=2, right=3, bottom=4, textalign=text.align_left)  // box uses text_halign

Three things make Pine unusually hostile to a language model:

1. Parameter names are not guessable. label.new takes textalign. box.new takes text_halign. Nothing about either name implies the other.
2. Several functions have two valid call forms. line.new, label.new and box.new each accept a chart.point or independent coordinates. A model that has only seen one form will insist the other is wrong.
3. The language keeps moving. TradingView shipped request.footprint(), multiline strings and calc_on_every_history_tick in the last eighteen months — and removed the wrapped-line indentation rules in December 2025. Training data goes stale in both directions.

The result is a loop you have probably lived: the agent writes Pine, you paste it into TradingView, it fails, you paste the error back, it guesses again.

What this does about it

Gives the agent a checker and a reference, so it stops guessing.

MCP tools

Tool	What it does
validate_pine_script	Runs the real validator, returns structured diagnostics with line numbers
lookup_pine_reference	The actual signature of any v6 built-in — parameters, and every overload

Skills

Skill	Covers
pinescript-v6	Execution model, overloaded constructors, anti-repainting, platform limits, API through July 2026
pinescript-validation	Every diagnostic class and its deterministic fix — plus what the validator cannot see
pinescript-indicator	Plotting, drawing objects, tables, alerts. Three CI-validated scaffolds
pinescript-strategy	Entries, exits, risk sizing, the five traps that make a backtest lie, webhook payloads

Hook

Every .pine file the agent edits is validated automatically. Not "remember to check your work" — an actual control.

The loop closes inside the conversation instead of across TradingView's compiler.

Two kinds of wrong

Pine breaks in two distinct ways, and most tooling conflates them.

Syntactic — hallucinated functions, wrong parameter names, v4/v5/v6 mixed together. Caused by a thin training corpus and a language that changes quarterly. Costs a minute. Solved here.

Semantic — code that compiles perfectly and is still wrong. A repainting signal. ta.* inside a conditional, silently corrupting its own history. An accumulator missing var, resetting every bar. Costs a funded account:

"one overlooked mistake — like a repainting signal or scope error — can invalidate months of backtesting" — PickMyTrade

Prose cannot fix semantic errors, because the author already believes they are right. Nine semantic checks are specified in SPEC.md and being implemented in the engine — repainting detection, ta.*-in-conditional, accumulator state, lazy-evaluation traps, and the platform limits TradingView enforces but no local tool counts.

Plugin, or VS Code extension?

Both — and the distinction matters, because it is what stops the two disagreeing.

        pinescript-v6-validator  (npm)     ← detection lives here, once
                     │
        ┌────────────┴────────────┐
        ▼                         ▼
  VS Code extension          this plugin
  surface: humans            surface: agents
  squiggles, hover           MCP, skills, hook

A check is written once, in the engine. The extension renders it as a squiggle; this plugin returns it to the model. Neither reimplements it — the moment the same rule exists twice they drift, and a drifted rule means your agent and your editor disagree about the same file.

Skills are plugin-only: prose is useless in an editor and is the whole point for an agent. Editor affordances — formatting, go-to-definition — stay in the extension.

Why this one

It shares an engine with a published VS Code extension. The validator here is the same one running in pinescript-vscode-extension — 1,400+ installs, 4.45★. So the agent and your editor cannot disagree about a file.

That engine carries:

• 457 function signatures scraped from the official v6 reference, plus a hand-maintained layer for everything TradingView shipped since
• Explicit overload modelling — a call is valid if it satisfies any form
• A golden corpus of scripts verified to compile on TradingView, asserted to produce zero errors on every commit

Every Pine example in every skill is extracted and run through that validator in CI. A skill shipping code that fails its own checker would make the whole thing worthless, so the build blocks it.

Install

/plugin marketplace add jpantsjoha/pinescript-plugin
/plugin install pinescript-plugin

The validation engine comes from pinescript-v6-validator on npm — no separate checkout needed.

For Claude Code:

/plugin marketplace add jpantsjoha/pinescript-plugin
/plugin install pinescript-plugin

It also packages for Antigravity/Gemini (gemini-extension.json + GEMINI.md), Codex (.agents/skills/ + AGENTS.md) and Kimi (.kimi-plugin/). All four manifests are checked for consistency by make gate.

To develop against an unreleased engine, point at a built checkout instead:

export PINESCRIPT_VALIDATOR=/path/to/pinescript-vscode-extension

Status: early. Four skills so far. See CHANGELOG.

What's in it

skills/pinescript-v6/          execution model, overloads, anti-repainting, limits
skills/pinescript-validation/  every diagnostic class and its deterministic fix
skills/pinescript-indicator/   validated scaffolds: overlay, oscillator, drawings
skills/pinescript-strategy/    entries, exits, sizing, and the repainting traps
mcp/server.js                  validate_pine_script + lookup_pine_reference
hooks/                         validates every .pine file the agent edits
scripts/                       spec, packaging, skill, link and example validation
tests/                         MCP behaviour tests

make gate runs everything: Agent Plugins 1.0.0 conformance, multi-client packaging consistency, skill frontmatter contracts, reference-URL resolution, embedded Pine validation, and the MCP behaviour tests.

Related projects

Project	What it is
pinescript-vscode-extension	The VS Code extension — IntelliSense, hover docs, real-time validation. Source of the engine this plugin uses.
googlecloud-plugin	Same plugin architecture applied to Google Cloud delivery — solution design, security, SRE, agentic patterns.

Prior art

TradersPost Pine Script plugin covers similar territory and is worth a look. The distinction claimed here is narrow and checkable: a shared validation engine with a published extension, and a CI gate that runs every documented example through it.

Contributing

Found Pine that this validator gets wrong? That is the most valuable bug report available — a false positive is worse than a missed error, because it puts red marks on correct code. Open an issue with the smallest script that reproduces it.

Author

Jaroslav Pantsjoha

• Website: jpantsjoha.com
• GitHub: @jpantsjoha
• LinkedIn: in/johas

License

MIT