What is plum CLAUDE.md?

Project instructions for Plum, a software project written in Go. They define checks to run before code is pushed and tasks for keeping marketplace information current.

Use them when working on Plum, especially before pushing changes or updating settings. They cover running the linter, tests, build, optional integration tests, and marketplace-data checks.

They help catch linting, test, and build failures before changes reach the main codebase. They also reduce the chance that project statistics and plugin counts become outdated.

plum CLAUDE.md — Instructions file by itsdevcoffee

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 instructions/itsdevcoffee/plum/claude-md

Clone the repo

git clone --depth 1 https://github.com/itsdevcoffee/plum

Per session 506 This file is loaded in full into every session.

When invoked 506 The same file — it is already loaded in full.

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	$0.00506	$0.00506
Opus 5	$0.00253	$0.00253
Sonnet 5	$0.00101	$0.00101
Haiku 4.5	$0.00051	$0.00051

Measured 2d ago against content hash afb1157df8ec, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plum CLAUDE.md 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 2d 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.

CLAUDE.md · 70 lines

What it actually says

Plum Project Guide

Pre-Push Checklist

ALWAYS run before pushing to main:

# 1. Run linter (must pass)
golangci-lint run --timeout=5m

# 2. Run tests (must pass)
go test ./...

# 3. Verify build succeeds
go build -o ./plum ./cmd/plum

# 4. Run integration tests (optional but recommended for settings changes)
go test -tags=integration ./internal/integration/... -v

If linter is not installed:

go install github.com/golangci/golangci-lint/v2/cmd/[email protected]

Routine Maintenance

At the start of each session, check marketplace data freshness:

# 1. Update GitHub stats (stars, forks, last updated)
bash scripts/update-marketplace-stats.sh

# 2. Check plugin counts from live manifests
go run scripts/check-plugin-counts.go

If changes detected:

Update internal/marketplace/discovery.go with new stats and plugin counts
Update README.md marketplace table with accurate counts
Update total plugin count in README intro and features section
Run pre-push checklist above

Why This Matters

Linting - CI will fail if linting doesn't pass locally. Always lint before pushing.
Accurate Data - The README is users' first impression. Accurate plugin counts and GitHub stats help them make informed decisions about installation.

Documentation

Files go in docs/ except for obvious exceptions: README.md, CLAUDE.md, LICENSE.md, CONTRIBUTING.md, AGENT.md, ... (root only).

Subdirectories:

context/ - Architecture, domain knowledge, static reference
decisions/ - Architecture Decision Records (ADRs)
handoff/ - Session state for development continuity
project/ - Planning: todos, features, roadmap
research/ - Explorations, comparisons, technical analysis
tmp/ - Scratch files (safe to delete)

Naming: YYYY-MM-DD-descriptive-name.md (lowercase, hyphens)

Rules:

Update existing docs before creating new ones
Use tmp/ when uncertain, flag for review

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.

2d ago First seen · 70 lines · 506 tokens per session scan A afb1157df8ec

Subscribe to this mod's changes

plum CLAUDE.md is an instructions file published in the GitHub repository itsdevcoffee/plum (11 stars, last pushed 6mo ago), licensed MIT. It adds 506 tokens to every session, about $0.0025 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.

Languages Testing bubbletea claude claude-code cli command-line marketplace

plum CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

Plum Project Guide

Pre-Push Checklist

Routine Maintenance

Why This Matters

Documentation

What this file has done since we first saw it

Other instructions, from other repositories

vue-tui AGENTS.md

axe AGENTS.md

UniCli CLAUDE.md

autospec AGENTS.md

killpy AGENTS.md

tui-design-skill AGENTS.md