spec

spec is a command for coding agents from Smart-AI-Memory/attune-ai. It costs 20 tokens per session (451 once invoked), scanned A, original, Apache-2.0.

A spec-driven development workflow that helps create, review, import, and execute detailed implementation plans with approval and quality checks.

In plain words
What is it for?
Use it to turn an idea into tasks, resume or import a plan, review its risks, and implement approved tasks one by one.
Why use it?
It gives complex work a saved plan and checks each task before moving on, reducing the chance of missed requirements or unchecked changes.

Command

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 commands/smart-ai-memory/attune-ai/spec
Clone the repo
git clone --depth 1 https://github.com/Smart-AI-Memory/attune-ai

Wrote 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.

agentmods badge for spec

README.md
[![agentmods](https://agentmods.dev/badge/commands/smart-ai-memory/attune-ai/spec.svg)](https://agentmods.dev/commands/smart-ai-memory/attune-ai/spec)
Your own site
<a href="https://agentmods.dev/commands/smart-ai-memory/attune-ai/spec"><img src="https://agentmods.dev/badge/commands/smart-ai-memory/attune-ai/spec.svg" alt="Measured on agentmods" height="20"></a>
Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 451 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 $0.00020 $0.00451
Opus 5 $0.00010 $0.00226
Sonnet 5 $0.00004 $0.00090
Haiku 4.5 $0.00002 $0.00045

Measured yesterday against content hash 767e1e86eb73, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spec 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.

src/attune/commands/spec.md · 62 lines

What it actually says

Spec-driven development for $ARGUMENTS.

If no arguments, use AskUserQuestion to ask what the user wants to do:

  • Start a new spec (brainstorm, decompose, save)
  • Resume an in-progress spec (check for resumable plans)
  • Import a spec file (load from another project/path)
  • Execute a spec (review then task-by-task with approval)

Import

If the user provides a file path or chooses "Import":

  1. Validate path with _validate_file_path()
  2. Copy to .claude/plans/ if not already there
  3. Load tasks with read_spec(path)
  4. If tasks found, proceed to Review
  5. If no tasks, offer to create a spec instead

Create

Run brainstorm flow (Context, Problem, Goals, End State). Auto-decompose approach into XML <task> blocks. Save to .claude/plans/.

Review

Load tasks with read_spec(plan_path). Show each task: name, objective, files, risks. Use AskUserQuestion for approve/edit/reject.

Execute

For each pending task:

  1. Show progress bar
  2. Show task detail
  3. Implement the task (create/modify files)
  4. Run quality gates via PipelineOrchestrator
  5. Severity-gated approval:
    • HIGH severity (score < 50): only "Fix and retry" or "Acknowledge risk" (no auto-run)
    • MEDIUM/LOW: "Approve" / "Redo" / "Auto-run remaining"
  6. Save state after each decision

Resume

Check find_resumable_plans() on startup. Offer to resume incomplete specs.

Use from attune.spec import for all helpers: present_tasks, present_task_detail, format_progress_bar, load_state, save_state, find_resumable_plans, get_pending_tasks, SpecState.

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 · 62 lines · 20 tokens per session scan A 767e1e86eb73

Subscribe to this mod's changes

spec is a command published in the GitHub repository Smart-AI-Memory/attune-ai (10 stars, last pushed today), licensed Apache-2.0. It adds 20 tokens to every session and 451 once invoked, about $0.0001 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 2026-09-03.