# Cococonverge

CocoPivot, multi-pod convergence synthesizer.

- Repository: https://github.com/Snowflake-Labs/cocoplus (727 stars)
- Category: Engineering
- License: MIT
- Last updated: 2026-10-04
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add Snowflake-Labs/cocoplus --skill cococonverge -g -a claude-code -y`
Codex: `npx skills add Snowflake-Labs/cocoplus --skill cococonverge -g -a codex -y`
Cursor: `npx skills add Snowflake-Labs/cocoplus --skill cococonverge -g -a cursor -y`
Gemini CLI: `npx skills add Snowflake-Labs/cocoplus --skill cococonverge -g -a gemini-cli -y`

## Description

CocoPivot, multi-pod convergence synthesizer. Deterministically deduplicates findings from N parallel pod outputs into one prioritized, coordinated action plan. Invoked via $pivot run/show/status/clear.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

## Objective

You are executing a CocoPivot command. When N CocoPods inspect the same codebase in parallel, they produce N separate findings lists, often the same underlying issue described three different ways from three different lenses. CocoPivot's discipline is to be the **sole author of `lifecycle/FINDINGS.md`**, so the developer has exactly one authoritative source for what needs addressing and in what order. Without CocoPivot, parallel pods produce parallel noise. With it, they produce coordinated signal.

Before proceeding, verify that `.cocoplus/` exists. If not, output: "CocoPlus is not initialized. Run `$pod init` first." Then stop.

## Commands

### `$pivot run`

Use the `cococonverge/pivot-merge` contract in `run` mode. Read all pod output files produced by the most recent CocoFlow `parallel:` step execution (from `pod-status.json`), apply the three-pass deduplication algorithm below, assign priority tiers and effort estimates, and write `lifecycle/FINDINGS.md` and `lifecycle/findings-state.json`. If the upstream flow used `on_partial: skip_partial`, exclude PARTIAL pod outputs.

### `$pivot run --since <timestamp>`

Use the `cococonverge/pivot-merge` contract in `run` mode and target only pod outputs that completed after `<timestamp>`. Useful for re-running convergence when additional pods have completed since the last run.

### `$pivot show`

Display the current `lifecycle/FINDINGS.md`, grouped by priority tier. Include the Coverage Note section if any contributing pod was PARTIAL.

### `$pivot status`

Read `lifecycle/findings-state.json` and display: total unique findings, count by priority tier (P1/P2/P3/P4), count by severity (BLOCKING/IMPORTANT/MINOR/ADVISORY), list of contributing pods with completion status (COMPLETE/PARTIAL/ERROR), and timestamp of the last convergence run.

### `$pivot clear`

The rest: https://github.com/Snowflake-Labs/cocoplus/blob/HEAD/.cortex/skills/cococonverge/SKILL.md