prime

prime is a command for Claude Code from dsifry/metaswarm. It costs 10 tokens per session (1,544 once invoked), scanned A, original, MIT.

A command that loads relevant facts from a project knowledge base into the agent's working context before it investigates, plans, implements, or reviews code.

In plain words
What is it for?
Use it when starting an issue, changing code areas, or switching tasks. It can gather context automatically or by selected files, keywords, or work type.
Why use it?
It helps the agent follow existing project rules and patterns, avoid known pitfalls, and not repeat previously learned mistakes.

Command for Claude Code

Part of the metaswarm plugin — 14 skills, 16 commands, 19 agents, 2 hooks shipped together

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/dsifry/metaswarm/prime
Clone the repo
git clone --depth 1 https://github.com/dsifry/metaswarm

Made for: Claude Code.

Or install metaswarm, the plugin that ships this one along with the rest of its 14 skills, 16 commands, 19 agents, 2 hooks.

Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,544 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 $0.00010 $0.01544
Opus 5 $0.00005 $0.00772
Sonnet 5 $0.00002 $0.00309
Haiku 4.5 $0.00001 $0.00154

Measured 3d ago against content hash 8fc65af9e932, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prime 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 3d 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/commands/prime.md · 260 lines

How it starts

The opening of the file, as written. The whole thing — 260 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BEADS Prime

CRITICAL: Run this command at the START of any investigation, planning, or implementation work to load relevant knowledge into your context.

When to Use

  • Starting work on a GitHub Issue
  • Beginning investigation/research
  • Before writing a plan
  • Before implementing changes
  • When switching to a new area of the codebase

How It Works

This command queries the BEADS knowledge base for facts relevant to your current context and injects them into the conversation, ensuring you:

  1. Follow established patterns and rules
  2. Avoid known gotchas and pitfalls
  3. Make decisions aligned with architectural choices
  4. Don't repeat mistakes that have been learned from

Usage

Quick Prime (Most Common)

For general context with automatic detection:

bd prime

Prime for Specific Files

When working on specific files:

bd prime --files "src/lib/services/**/*.ts" "src/api/routes/**/*.ts"

Prime for Keywords

When working on a specific topic:

bd prime --keywords "authentication" "jwt" "security"

Prime for Work Type

When doing a specific type of work:

bd prime --work-type planning
bd prime --work-type implementation
bd prime --work-type review
bd prime --work-type debugging

Prime for Context Recovery

When resuming after context compaction or in a new session:

bd prime --work-type recovery

This loads active plan, project context, execution state, and relevant knowledge base facts. See "Context Recovery" section below.

Combined

bd prime \
  --files "src/lib/services/ai-*.ts" \
  --keywords "ai" "provider" "openai" \
  --work-type implementation

What Gets Loaded

1. MUST FOLLOW (Critical Rules)

Non-negotiable rules containing NEVER/ALWAYS/MUST:

  • "NEVER use as any type casting"
  • "ALWAYS use centralized AI config"
  • Security-critical patterns

2. GOTCHAS (Common Pitfalls)

Known issues to avoid:

  • "Truthy check fails for explicit zero values - use !== undefined"
  • API behavior quirks

Read the full file on GitHub · 260 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.

  1. 3d ago First seen · 260 lines · 10 tokens per session scan A 8fc65af9e932

Subscribe to this mod's changes

prime is a command published in the GitHub repository dsifry/metaswarm (407 stars, last pushed 2mo ago), licensed MIT. It adds 10 tokens to every session and 1,544 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 2026-08-30.