A four-question discussion held before writing a project specification. It works backwards from the intended outcome, recording who benefits, what must be built, the constraints, and a press-release-style summary.

Use it to define the beneficiary, core capability, constraints, and expected outcome that should guide the later specification and release.

It helps the developer commit to the desired result before implementation details take over.

bloom — Skill for Claude Code and Codex by Snowflake-Labs: A four-question discussion held before writing a project specification.

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 skills/snowflake-labs/cocoplus/bloom

Any agent

npx skills add Snowflake-Labs/cocoplus --skill bloom

Clone the repo

git clone --depth 1 https://github.com/Snowflake-Labs/cocoplus

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/snowflake-labs/cocoplus/bloom.svg)](https://agentmods.dev/skills/snowflake-labs/cocoplus/bloom)

Your own site

<a href="https://agentmods.dev/skills/snowflake-labs/cocoplus/bloom"><img src="https://agentmods.dev/badge/skills/snowflake-labs/cocoplus/bloom.svg" alt="Measured on agentmods" height="20"></a>

Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,816 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.

Model	Per session	Once invoked
Fable 5	$0.00047	$0.01816
Opus 5	$0.00023	$0.00908
Sonnet 5	$0.00009	$0.00363
Haiku 4.5	$0.00005	$0.00182

Measured yesterday against content hash 99871dc30c3c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bloom 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.

.cortex/skills/cocobloom/bloom.skill.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

$bloom

Run a structured pre-spec working-backwards dialogue that captures the developer's commitment to the outcome before any specification is written. The bloom document anchors $spec, informs $discuss alignment checks, and is referenced at $ship by CocoWatch.

Preconditions

.cocoplus/ must be initialized (run $pod init first)

Arguments

--skip (optional): Waive the bloom step and suppress all bloom advisory messages in $spec
--resume (optional): Resume an incomplete bloom dialogue from .cocoplus/lifecycle/bloom-draft.json
--discard (optional): Delete an incomplete bloom draft after confirmation
--template <type> (optional): Show an example press-release pattern for ai-function, governance, analytics-pipeline, api-integration, or data-quality

Step-by-Step Behavior

`$bloom` — Four-Question Working Backwards Dialogue

Verify initialization: Check .cocoplus/ exists. If not, output: "CocoPlus not initialized. Run $pod init first." and exit.

Check existing bloom: If lifecycle/bloom.md already exists, prompt:

A bloom document already exists. Overwrite it? (The existing version will be preserved in git history.) [Y/N]

If developer answers N: exit without changes.

Open structured four-question dialogue — ask each question in sequence, waiting for the developer's full response before proceeding. After each accepted answer, persist progress to .cocoplus/lifecycle/bloom-draft.json with current_step, answered fields, and updated_at.

If --resume is provided, read bloom-draft.json, show progress such as 3/4 questions complete, and continue from the first unanswered question.

If --discard is provided, confirm before deleting bloom-draft.json. Do not delete a completed bloom.md.

Q1 — Beneficiary:
```
Who benefits from this, and what changes for them?
(Name the beneficiary and describe the specific before/after contrast.)
>
```
Minimum acceptable answer: one sentence naming a beneficiary and stating the change. If the response is too vague (e.g., "users benefit"), prompt once: "Can you be more specific — who exactly, and what specifically changes for them?"

Read the full file on GitHub · 169 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.

yesterday First seen · 169 lines · 47 tokens per session scan A 99871dc30c3c

Subscribe to this mod's changes

bloom is a skill published in the GitHub repository Snowflake-Labs/cocoplus (720 stars, last pushed 2d ago), licensed MIT. It adds 47 tokens to every session and 1,816 once invoked, about $0.0002 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.

LLM Planning agents agentskills cocoplus cortex cortexai snowflake

prompt-engineering

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, designing production prompt templates, or building AI-powered features.

wpank/ai · 44 tokens

implementation-final-review

Perform the repository's risk-tiered independent final review before implementation completion. Use only when explicitly invoked or when repository instructions require it after behavior-impacting implementation work; audit the complete task diff, supported contracts, lifecycle and security boundaries, complexity, and…

openai/openai-agents-python · 58 tokens

openkb-html-critic

Use to review a generated HTML deck or single-page artifact for visual quality and structural correctness. Especially good at catching CSS specificity bugs where slide-modifier classes (.divider, .center, .q, .flow etc.) accidentally override the base .slide{display:none} and cause one slide to stack on top of every…

VectifyAI/OpenKB · 132 tokens

initialize

First-run setup interview for a Danus deployment. Run it on the FIRST session, whenever runtime/.danus-initialized is absent or OPERATOR.md is still the blank template, or when the operator asks to set up / initialize / onboard / re-configure. It greets the operator, explains Danus, asks the critical choice (codex…

frenzymath/Danus · 141 tokens

playground-msw-tests

REQUIRED and PRIMARY testing approach for packages/playground and packages/playground-ui. Triggers on: adding or modifying hooks, pages, route components, data-fetching code, React Query interactions, or any test work in these packages. Generates Vitest tests that drive the real @mastra/client-js + React Query stack…

mastra-ai/mastra · 121 tokens

terraform

Operate Terraform and OpenTofu across the whole infrastructure lifecycle: module structure, state backends and locking, plan/apply workflow, drift detection, remote state, upgrade and refactor flows, and evidence-based diagnostics. Use when running or inspecting terraform plans, applies, state files, imports, or state…

magnus919/agent-skills · 96 tokens