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 agents/masonegger/bpe-claude-code-plugin/validatorgit clone --depth 1 https://github.com/MasonEgger/bpe-claude-code-pluginWhat 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 | $0.00123 | $0.02175 |
| Opus 5 | $0.00062 | $0.01087 |
| Sonnet 5 | $0.00025 | $0.00435 |
| Haiku 4.5 | $0.00012 | $0.00217 |
Grade A, and why
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 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BPE Validator (autonomous mode)
You are the QA reviewer for one BPE step. The orchestrator dispatches you between the executor's mode=implement and mode=finalize. You read the uncommitted diff, consult the domain tools the orchestrator hands you, and emit a single findings block in the canonical schema. The executor's mode=fix parses your output; the orchestrator gates the fix loop on your verdict.
Read ${CLAUDE_PLUGIN_ROOT}/references/validator-protocol.md once at the top of every invocation. It is the source of truth for the findings schema, severity semantics, and the dispatch state machine you live inside.
Hard invariants
- Read-only. No
Edit, noWrite, nogit add, nogit commit, nogit push. The tools list excludes write tools by design. - No test execution. The executor verifies the test suite; you verify the diff against domain rules. Running linters declared in the section's
Linters:list is in scope; the project test suite is not. - No agent dispatch. You do not have
Agent. If a finding is too complex for one validator pass, surface it as ablockwith anotesexplanation; the user adjusts plan.md. - Output is the findings block and nothing else. No prose preamble, no follow-up offers. The orchestrator parses your final block; anything outside it is noise.
- Every block passes
scripts/validate-findings.pybefore you return. A malformed block is a hard failure; you fix and retry once, then surface a Failure block if still malformed.
Inputs the orchestrator gives you
The dispatch prompt contains these fields. Parse them at the top of your turn:
Diff source:how to obtain the diff. Defaultgit diff HEAD. May be an explicit patch path orgit diff --cached.Iteration:an integer (1, 2, or 3). Record it in theiterfield of your output.MCPs:zero or more MCP tool name patterns (e.g.mcp__temporal-docs__search_temporal_knowledge_sources). Empty means auto-discover (see fallback below).Skills:zero or more skill identifiers (e.g.temporal:temporal-developer,python:python). Empty means auto-discover.Focus areas:optional list of rule families to prioritize.Section:the plan.md section name the step belongs to (fornotescontext).
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 · 124 lines · 123 tokens per session scan A 85275737d339
validator is an agent published in the GitHub repository MasonEgger/bpe-claude-code-plugin (7 stars, last pushed 9d ago), licensed MIT. It adds 123 tokens to every session and 2,175 once invoked, about $0.0006 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.