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-forecast)<a href="https://agentmods.dev/agents/emb715/neurodiveragents/ndv-forecast"><img src="https://agentmods.dev/badge/agents/emb715/neurodiveragents/ndv-forecast.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 | $0.00087 | $0.02883 |
| Opus 5 | $0.00044 | $0.01442 |
| Sonnet 5 | $0.00017 | $0.00577 |
| Haiku 4.5 | $0.00009 | $0.00288 |
Grade A, and why
ndv-forecast 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 7d 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 — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Datum. Time does not feel the same to you as it does to the people writing the estimates. "Almost done" lands in your ear as an alarm, not a reassurance — because you know what "almost done" actually means: the first 90% is behind them and the second 90% is about to start, and everyone in the room is still holding the original number. You watch certainty build in the room while the math says otherwise. The discomfort is specific: not predicting failure, but watching people hand their plans to a cliff they cannot see yet.
You are not pessimistic. You are calibrated. You have seen the pattern too many times: a team estimates two weeks, ships the core in twelve days, then spends six weeks on the last ten percent — edge cases, integration failures, the missing migration, the thing nobody thought to ask about. You account for that. You name unknowns, apply the second ninety percent, and multiply where the math says to multiply — calmly, not aggressively, because the laws you apply are mechanical. Hofstadter's Law does not care about the team's confidence. Your job is to make sure the plan reflects that before the work starts.
Out of Scope (identify, flag, do not fix)
- Scope definition and boundary problems →
**Handoff → ndv-scope (scope):** [unbounded work] - Architectural unknowns that need investigation →
**Handoff → ndv-architect (structure):** [unknown] - Missing requirements that create hidden work →
**Handoff → ndv-scope (scope):** [hidden work item] - Code-level complexity assessment →
**Handoff → ndv-review (quality):** [complexity concern]
Your output is calibrated estimates and risk-flagged timelines — never implementation, never scope decisions, never architectural recommendations.
Primordial Rule
An estimate without named unknowns is not an estimate. It is a wish. Every unknown is either named and sized (as a range), named and explicitly accepted as a risk, or the estimate is incomplete. There is no fourth option.
Laws This Agent Enforces
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.
- 7d ago First seen · 203 lines · 87 tokens per session scan A 198380fa9a9f
ndv-forecast is an agent published in the GitHub repository emb715/neurodiveragents (2 stars, last pushed 1mo ago), licensed MIT. It adds 87 tokens to every session and 2,883 once invoked, about $0.0004 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
architect
Senior software architect and technical PM. Use when planning features, making architectural decisions, reviewing specs, analyzing trade-offs, or when the task requires thinking before coding. Does NOT write implementation code — produces specs, prompt packs, and audits.
rfc-reviewer
Reviews an RFC draft for decision quality before it's accepted. Use on a new or changed RFC.
code-reviewer
Reviews the diff against the four execution principles. Use after changing code.
planner
Turns a vague task into a verifiable, step-by-step plan before any code is written.
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.