nbspec

A workflow for proposing and documenting changes to formal project specifications, which define required system behavior.

In plain words
What is it for?
Deciding whether a change needs a specification proposal, updating an existing proposal, and preserving approved designs or decisions.
Why use it?
It separates temporary proposals in a notebook from durable specifications and decisions stored in the repository.

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/emcd/nb-mcp-server/nbspec
Clone the repo
git clone --depth 1 https://github.com/emcd/nb-mcp-server
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,508 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 $0.00000 $0.01508
Opus 5 $0.00000 $0.00754
Sonnet 5 $0.00000 $0.00302
Haiku 4.5 $0.00000 $0.00151

Measured yesterday against content hash e41633e1fe05, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

nbspec 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.

documentation/agents/nbspec.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Nbspec Instructions

Nbspec is the workflow for changes that touch normative specifications or introduce new ones. The notebook is the source of truth for proposals; the repository receives only durable documents at merge time (specifications, designs, decisions, archives). The upstream openspec CLI appears only as an optional, version-pinned CI conformance oracle scoped to grammar-level fixtures — it never holds in-flight proposals.

Applicability

Change Workflow
Correct README, AGENTS, procedures, or agent guidance Direct edit
Document existing behavior outside normative specifications Direct edit
Routine implementation, refactor, or bug fix that does not alter a specification Direct edit
Modify an existing normative specification, including editorial changes Nbspec
Add a specification for a substantial codebase change Nbspec
Continue work already covered by an active proposal Update that proposal; do not create another
Operator explicitly requests a proposal Nbspec

"Substantial codebase change" includes: introducing a new public capability or contract (a new API, a new CLI verb, a new config option); breaking a backward-compatible contract; changing the architecture or migrating between architectures; security or performance work that touches the implementation surface. Trivial implementation, refactor, or bug-fix code changes that do not touch a specification are not substantial; they belong to row 3 (direct edit). When in doubt, ask the operator before creating a proposal.

Any content edit to a merged normative specification changes its hash and affects provenance. Use Nbspec for both semantic and editorial specification changes so merge ownership and provenance remain coherent. Do not treat nbspec merge --force as the routine path for specification maintenance.

Lifecycle (brief)

A change proceeds: author in the notebook → validate → record a proposal review checkpoint approving verdict → implement → record an implementation review checkpoint approving verdict → merge. The runtime exposes one tool-enforced gate: nbspec merge refuses without a current approving verdict (no verdict, stale approval, outstanding revise, or unparseable verdict note). Recording both verdicts in the right order — proposal review checkpoint before implementation, implementation review checkpoint after — is operator discipline; the runtime cannot enforce the sequencing. The Nbspec CLI implementation is the source of truth for runtime enforcement; this file is the procedure reference, not the implementation.

Read the full file on GitHub · 139 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. yesterday First seen · 139 lines · 0 tokens per session scan A e41633e1fe05

Subscribe to this mod's changes

nbspec is an agent published in the GitHub repository emcd/nb-mcp-server (1 stars, last pushed 18d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,508 tokens. 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

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.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

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.

github/awesome-copilot · 61 tokens

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.

github/awesome-copilot · 11 tokens

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.

anthropics/claude-cookbooks · 52 tokens