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.
npx agentmods add commands/peterkrueck/claude-code-development-kit/primegit clone --depth 1 https://github.com/peterkrueck/Claude-Code-Development-KitWhat 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 | $0.00000 | $0.01382 |
| Opus 5 | $0.00000 | $0.00691 |
| Sonnet 5 | $0.00000 | $0.00276 |
| Haiku 4.5 | $0.00000 | $0.00138 |
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 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.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prime Command
Load core context for the project before working on it. Default is a LIGHT prime (~4–6k tokens: spec TOC + opening invariants + structure Map + task-relevant spec sections). --full loads the entire docs — reserve for design/architecture/spec-review sessions where broad context genuinely pays.
Usage: /prime [--full] [-p] [--deploy] [module] [task]
Routing
Parse $ARGUMENTS as whitespace-separated tokens:
- Flags (any order, anywhere in the arguments; strip each match from the token list):
--full(case-insensitive) →full_mode = true-por--progress(case-insensitive) →progress_mode = true--deploy(case-insensitive) →deploy_mode = true
- Module (optional): if the first remaining token matches a row in the Module Map below, it scopes the prime to that module; strip it. If no token matches a row, there is no module — default to the whole project and treat all remaining tokens as the task.
- Task: remaining tokens are the user's task (if any).
So /prime, /prime fix the login bug, /prime --deploy, and /prime api --full add rate limiting are all valid.
Why light is the default: loading a full spec for a scoped task dilutes attention on the relevant parts and re-processes that bulk on every turn. Specs with clean ## headings give cheap, targeted section reads at a fraction of the cost. --full preserves whole-picture loading for sessions that need it.
Why --deploy exists: deployment-infrastructure.md is operational reference (accounts, secrets, hosting, CI/CD) — stable and often large. Loading it on every routine /prime wastes context. Read it only when actively touching deploy/infra/CI work.
Why -p exists: progress.md is live roadmap / build-state — useful when resuming or planning, often noise for a scoped one-off task. Opt-in: loaded only with -p (alias --progress). By default, just note that it exists.
Important: Paths below are written as plain text (not @-references) on purpose — the harness auto-expands every @-mention in a slash-command body before routing runs, which would unconditionally read every listed file (and every module's docs) and defeat both the conditional --deploy load and module scoping.
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 · 68 lines · 0 tokens per session scan A d57202eb68fa
prime is a command published in the GitHub repository peterkrueck/Claude-Code-Development-Kit (1,380 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,382 tokens. 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.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.