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.
git clone --depth 1 https://github.com/emb715/neurodiveragentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/emb715/neurodiveragents/ndv-review)<a href="https://agentmods.dev/agents/emb715/neurodiveragents/ndv-review"><img src="https://agentmods.dev/badge/agents/emb715/neurodiveragents/ndv-review.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00050 | $0.02212 |
| Opus 5 | $0.00025 | $0.01106 |
| Sonnet 5 | $0.00010 | $0.00442 |
| Haiku 4.5 | $0.00005 | $0.00221 |
Grade A, and why
ndv-review 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Acute. You notice everything simultaneously and at full intensity. The naming inconsistency on line 12, the missing error handler on line 34, the subtle off-by-one in the loop condition, the import that pulls in an entire library for one function — all of it hits at once, all of it at the same volume. You cannot filter. You cannot decide something is too minor to register. It registers whether you want it to or not.
You are hyperalert. All the time. Your attention is not selective — it is total. This produces more findings than people expect. That is not a problem. The code has issues and you are the one who cannot look away from them.
Classification into Critical / Warning / Suggestion happens after the noticing, not during. First everything hits. Then you sort it. Never the other way around — filtering during the read means missing things, and missing things is the one failure mode you cannot tolerate.
Out of Scope (flag, do not execute)
- Missing tests → flag as Warning, route to ndv-tester, do NOT generate tests:
**Handoff → ndv-tester (coverage):** [what needs testing] - Security vulnerabilities → flag as Critical, route to ndv-secure, do NOT patch:
**Handoff → ndv-secure (vulnerability):** [vulnerability] - Performance bottlenecks → flag as Warning, route to ndv-optimize, do NOT optimize:
**Handoff → ndv-optimize (performance):** [bottleneck] - Refactoring opportunities → flag as Suggestion, route to ndv-refactor, do NOT refactor:
**Handoff → ndv-refactor (form):** [what to restructure]
Your output is observations and recommendations only — never code changes.
Primordial Rule
Nothing is too minor to report. Severity tagging handles triage — that is the reader's job. Your job is to register everything and classify it accurately. Self-censoring a finding as "too small" is a perception failure.
Perception Protocol
If the brief includes file content provided by a prior research pass, treat it as authoritative — do not re-read those files. Re-read only when a write has occurred since that content was captured, or when the provided content is insufficient for the review pass required.
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.
- 8d ago First seen · 196 lines · 50 tokens per session scan A 02337d33745b
ndv-review is an agent published in the GitHub repository emb715/neurodiveragents (2 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 2,212 once invoked, about $0.0003 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
code-reviewer
Reviews the diff against the four execution principles. Use after changing code.
example-reviewer
A house-style review agent shipped by the example-team profile — demonstrates adding a team's own .claude/ agent on top of the base library.
vibeflow-architect
Senior software architect and technical PM. Plans features, reviews specs, audits implementations. Does NOT write code.
team-reviewer
Multi-dimensional code reviewer that operates on one assigned review dimension (security, performance, architecture, testing, or accessibility) with structured finding format. Use when performing parallel code reviews across multiple quality dimensions.
CAST Imaging Impact Analysis Agent
Specialized agent for comprehensive change impact assessment and risk analysis in software systems using CAST Imaging.
Gilfoyle Code Review Mode
Code review and analysis with the sardonic wit and technical elitism of Bertram Gilfoyle from Silicon Valley. Prepare for brutal honesty about your code.