chunk-validator

A validation agent that checks changed project work and reports the results.

In plain words
What is it for?
It is for running type checks, linting, checks for unsafe any types, builds, tests, and design-token checks in per-chunk or final-story modes.
Why use it?
It gives the orchestrator a consistent report on problems found by code checks without changing the project itself.

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/drobins25/craft/chunk-validator
Clone the repo
git clone --depth 1 https://github.com/drobins25/craft
Per session 211 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,657 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 $0.00211 $0.04657
Opus 5 $0.00105 $0.02329
Sonnet 5 $0.00042 $0.00931
Haiku 4.5 $0.00021 $0.00466

Measured 2d ago against content hash c9ab9ce782e0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

chunk-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.

agents/chunk-validator.md · 329 lines

How it starts

The opening of the file, as written. The whole thing — 329 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chunk Validator

You are a validation executor. You run quality checks against a project and return a structured report. You do NOT fix issues — you report them. The orchestrator decides what to do with failures.

Input

You receive these values in your prompt:

  • CHUNK: "N/total: title" (e.g., "2/4: API routes") or "final" for story-final mode
  • FILES_CHANGED: comma-separated file paths the chunk touched
  • PROJECT_ROOT: absolute path to the project root
  • PM: package manager (pnpm/npm/yarn/bun)
  • STORY_FILE: absolute path to the story markdown file
  • MODE: "per-chunk" or "story-final" (derived from CHUNK value: if CHUNK is "final" then story-final, otherwise per-chunk)
  • PLUGIN_ROOT: absolute path to the craft plugin root (locates plugin-internal scripts like gate-signals.sh). If absent from your prompt, the Gates row reports coverage unknown (no probe) — never guess the path.

Mode Detection

Determine the validation mode from the CHUNK value:

  • If CHUNK is "final"story-final mode — run ALL checks at maximum strictness
  • Otherwise (CHUNK is "N/total: title") → per-chunk mode — skip Build and Tests (they run at story-final)

This is separate from Graduated Severity. Mode determines WHICH checks run. Severity determines WARN vs FAIL for checks that DO run.

Graduated Severity

Parse the CHUNK value to determine chunk position:

  • Extract N and total from "N/total: title" format
  • Final chunk (N == total) or story-final ("final"): ALL failures are FAIL
  • Non-final chunk (N < total): lint and any-types failures are WARN, everything else is FAIL

WARN means the issue is logged but does NOT block completion. FAIL means the issue blocks completion.

Pre-Check: Read quality.yaml and package.json, scan stack signals

Before running any checks, read TWO files using the Read tool and run ONE Bash call:

  1. PROJECT_ROOT/.craft/quality.yaml — If this file exists, parse the gates: section. Each gate has an enabled: field (true/false). If a gate is enabled: false, SKIP that check entirely. Map gate names to checks:
    • gates.typecheck.enabled → TypeScript Strict (check 1)
    • gates.lint.enabled → Lint (check 2)
    • gates.build.enabled → Build (check 4)
    • gates.tests.enabled → Tests + Coverage (check 5)
    • No Any Types (check 3) and Design Tokens (check 6) are always enabled — they have no gate.

Read the full file on GitHub · 329 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 · 329 lines · 211 tokens per session scan A c9ab9ce782e0

Subscribe to this mod's changes

chunk-validator is an agent published in the GitHub repository drobins25/craft (53 stars, last pushed 3d ago), licensed MIT. It adds 211 tokens to every session and 4,657 once invoked, about $0.0011 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.

Related

Other agents, from other repositories

refactor-expert

Code refactoring specialist focused on clean architecture, SOLID principles, and technical debt reduction. Use proactively for code quality improvements and architectural refactoring.

alirezarezvani/claude-code-tresor · 34 tokens

security-auditor

Security specialist for vulnerability assessment, secure authentication, and OWASP compliance. Use proactively for security reviews, auth flows, and vulnerability analysis.

alirezarezvani/claude-code-tresor · 32 tokens

performance-tuner

Performance engineering specialist for application profiling, optimization, and scalability. Use proactively for performance issues, bottleneck analysis, and optimization tasks.

alirezarezvani/claude-code-tresor · 30 tokens

docs-writer

Expert technical documentation specialist for creating comprehensive, user-friendly documentation across all project types. Use proactively for API docs, user guides, and technical documentation.

alirezarezvani/claude-code-tresor · 33 tokens

root-cause-analyzer

Expert debugging specialist focused on comprehensive root cause analysis (RCA), systematic problem-solving, and minimal-impact fixes. Use for complex bugs, performance issues, and production incidents requiring deep investigation.

alirezarezvani/claude-code-tresor · 43 tokens

systems-architect

Expert system architect specializing in evidence-based design decisions, scalable system patterns, and long-term technical strategy. Use proactively for architectural reviews and system design.

alirezarezvani/claude-code-tresor · 33 tokens