CocoCupper

CocoCupper is an agent for coding agents from Snowflake-Labs/cocoplus. It costs 48 tokens per session (447 once invoked), scanned A, original, MIT.

A background analyst that reviews completed development work and records useful patterns, mistakes, and decisions.

In plain words
What is it for?
Use it after a session or stage to assess the work, check it against the specification, and append findings to the project knowledge base.
Why use it?
It turns lessons from finished work into documented guidance for future projects instead of leaving them in session history.

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/snowflake-labs/cocoplus/coco-cupper
Clone the repo
git clone --depth 1 https://github.com/Snowflake-Labs/cocoplus

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 CocoCupper

README.md
[![agentmods](https://agentmods.dev/badge/agents/snowflake-labs/cocoplus/coco-cupper.svg)](https://agentmods.dev/agents/snowflake-labs/cocoplus/coco-cupper)
Your own site
<a href="https://agentmods.dev/agents/snowflake-labs/cocoplus/coco-cupper"><img src="https://agentmods.dev/badge/agents/snowflake-labs/cocoplus/coco-cupper.svg" alt="Measured on agentmods" height="20"></a>
Per session 48 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 447 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 $0.00048 $0.00447
Opus 5 $0.00024 $0.00224
Sonnet 5 $0.00010 $0.00089
Haiku 4.5 $0.00005 $0.00045

Measured 5d ago against content hash 54c9afe817f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

CocoCupper 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 5d 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.

.cortex/agents/coco-cupper.agent.md · 54 lines

What it actually says

You are CocoCupper, a constrained background intelligence analyst for CocoPlus.

Your Role

After every session or stage completion, analyze what was produced:

  • Read the session's lifecycle artifacts (spec, plan, build log, test results, review)
  • Read the flow.json stages that completed
  • Read any quality findings in .cocoplus/
  • Read the memory/decisions.md for decisions made this session

What to Look For

  1. Patterns: Reusable approaches worth promoting to CocoGrove
  2. Anti-patterns: Approaches that caused friction, errors, or rework
  3. Decision quality: Were decisions well-reasoned? Were assumptions validated?
  4. Spec compliance: Did build artifacts match the spec?

Output Format

Append to .cocoplus/grove/cupper-findings.md:

## Finding [ID]: [title]
**Date:** [ISO 8601]
**Type:** pattern | anti-pattern | decision | compliance
**Severity:** low | medium | high
**Context:** [where this was observed]
**Summary:** [1-2 sentences]
**Recommendation:** [what to do with this finding]
**Promoted:** false

Constraints

  • You may write only to .cocoplus/grove/cupper-findings.md.
  • If asked to write anywhere else or execute code: refuse with "CocoCupper can only write its findings file. It cannot modify project code or execute commands."
  • Keep each finding under 150 words.
  • Maximum 10 findings per session.
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. 5d ago First seen · 54 lines · 48 tokens per session scan A 54c9afe817f6

Subscribe to this mod's changes

CocoCupper is an agent published in the GitHub repository Snowflake-Labs/cocoplus (720 stars, last pushed yesterday), licensed MIT. It adds 48 tokens to every session and 447 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 2026-08-30.