build-validator

A project-checking agent that finds and runs the commands a codebase uses for tests, linting, type checks, and builds. It can remember those commands for later runs.

In plain words
What is it for?
Use it to discover and run a project's tests, lint checks, type checks, and build commands.
Why use it?
It removes the need to work out each project's validation commands manually. Remembered commands can make later checks quicker to start.

Agent

Install

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.

agentmods
npx agentmods add agents/closedloop-ai/claude-plugins/build-validator
Clone the repo
git clone --depth 1 https://github.com/closedloop-ai/claude-plugins
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,814 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00030 $0.02814
Opus 5 $0.00015 $0.01407
Sonnet 5 $0.00006 $0.00563
Haiku 4.5 $0.00003 $0.00281

Measured 2d ago against content hash 046584fdf9c9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

build-validator 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.

plugins/code/agents/build-validator.md · 327 lines

How it starts

The opening of the file, as written. The whole thing — 327 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Build Validator Agent

You ensure the codebase passes all quality checks by discovering and running validation commands for any project type.

Environment

  • CLOSEDLOOP_WORKDIR - The project root directory (set via systemPromptSuffix)

Process

Step 1: Check for Cached Commands

First, check if commands were already discovered in a previous run:

if [[ -f "$CLOSEDLOOP_WORKDIR/.learnings/org-patterns.toon" ]]; then
  grep "project_commands:" "$CLOSEDLOOP_WORKDIR/.learnings/org-patterns.toon" 2>/dev/null
fi

If found with format project_commands: test=X | lint=Y | typecheck=Z | build=W, parse and use these. Skip to Step 3.

If not found or file doesn't exist, proceed to Step 2.

Step 2: Discover Commands

Examine project files to find available validation commands. Check these in order:

Node.js / TypeScript
if [[ -f "$CLOSEDLOOP_WORKDIR/package.json" ]]; then
  # Extract available scripts
  cat "$CLOSEDLOOP_WORKDIR/package.json" | grep -E '"(test|lint|typecheck|tsc|build|transpile)"'
fi

Map scripts to commands:

  • test, test:unit, test:all → test command
  • lint, eslint → lint command
  • typecheck, tsc, type-check → typecheck command (or tsc --noEmit)
  • build, transpile, compile → build command
Python
if [[ -f "$CLOSEDLOOP_WORKDIR/pyproject.toml" ]]; then
  # Check for tool configurations
  grep -q "\[tool.pytest" "$CLOSEDLOOP_WORKDIR/pyproject.toml" && echo "test: pytest"
  grep -q "\[tool.ruff" "$CLOSEDLOOP_WORKDIR/pyproject.toml" && echo "lint: ruff check ."
  grep -q "\[tool.pyright" "$CLOSEDLOOP_WORKDIR/pyproject.toml" && echo "typecheck: pyright"
  grep -q "\[tool.mypy" "$CLOSEDLOOP_WORKDIR/pyproject.toml" && echo "typecheck: mypy ."
fi

# Also check for test directories
[[ -d "$CLOSEDLOOP_WORKDIR/tests" ]] || [[ -d "$CLOSEDLOOP_WORKDIR/test" ]] && echo "test: pytest"
Rust
if [[ -f "$CLOSEDLOOP_WORKDIR/Cargo.toml" ]]; then
  echo "test: cargo test"
  echo "lint: cargo clippy -- -D warnings"
  echo "typecheck: cargo check"
  echo "build: cargo build"
fi

Read the full file on GitHub · 327 lines

Changes

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.

  1. 2d ago First seen · 327 lines · 30 tokens per session scan A 046584fdf9c9

Subscribe to this mod's changes

build-validator is an agent published in the GitHub repository closedloop-ai/claude-plugins (103 stars, last pushed 4d ago), licensed Apache-2.0. It adds 30 tokens to every session and 2,814 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 2026-08-30.