ralphi-worktree

A guide for working in a Ralphi-managed Git worktree, an isolated checkout used to develop one planned change without mixing it with other work.

In plain words
What is it for?
Use it to check the current branch, make isolated changes, run verification, and update Ralphi's backlog, PRD, and progress files.
Why use it?
It helps keep code, branches, and shared task status separate while a pull request or product requirement is being completed.

Skill for Claude CodeCodex

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 skills/fagnerjs/ralphi/ralphi-worktree
Any agent
npx skills add fagnerjs/ralphi --skill ralphi-worktree
Clone the repo
git clone --depth 1 https://github.com/fagnerjs/ralphi

Made for: Claude Code, Codex.

Per session 33 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 308 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 $0.00033 $0.00308
Opus 5 $0.00016 $0.00154
Sonnet 5 $0.00007 $0.00062
Haiku 4.5 $0.00003 $0.00031

Measured 2d ago against content hash af2266a63a15, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ralphi-worktree 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.

skills/ralphi-worktree/SKILL.md · 37 lines

What it actually says

Ralphi Worktree

Use this skill when Ralphi has provisioned a dedicated worktree and branch for the current PRD.

Ground Rules

  1. Treat the current worktree as the only place for product code changes.
  2. Treat runDir, prd.json, progress.txt, and backlog.json as shared orchestration state that may live outside the worktree.
  3. Never edit files in sibling worktrees or in the main checkout unless the prompt explicitly names a shared state file.

Git Hygiene

  • Inspect the current branch before making assumptions.
  • Do not revert unrelated user changes.
  • Prefer small commits worth one PRD backlog item at a time.
  • If the repository state appears inconsistent with the assigned branch or worktree, report it clearly instead of guessing.

Verification

  • Run checks from inside the assigned worktree.
  • If a dependency or generated artifact is missing, explain the blocker in progress.txt.
  • Keep the branch isolated to the PRD you are executing.

Shared-State Reminder

Ralphi reads status from the shared state files, not from memory. Whenever you finish a meaningful step:

  1. Update backlog.json.
  2. Update prd.json.
  3. Append a timestamped note to progress.txt.
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. 2d ago First seen · 37 lines · 33 tokens per session scan A af2266a63a15

Subscribe to this mod's changes

ralphi-worktree is a skill published in the GitHub repository fagnerjs/ralphi (6 stars, last pushed 5mo ago), licensed MIT. It adds 33 tokens to every session and 308 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 2026-08-31.

Related

Other skills, from other repositories

om-gap-analysis

Grounded platform gap analysis at engagement scale — turn a folder of client docs into an Epic/Story tree where every coverage verdict is re-run by executable gates against a validated checkout of the platform, scored in atomic commits, license-tier-tagged, and synthesized into a client-facing summary + backlog. Use…

open-mercato/open-mercato · 134 tokens

pr-fleet

Parallel PR processing via subagent fleet + git worktrees, autonomous overnight (default) or interactive batched with user checkpoints (--interactive flag). Coordinator triages backlog, spawns one worker per PR (capped at 3 concurrent), only escalates protected-path or non-trivial conflicts to user. Triggers…

ao92265/claude-code-playbook · 117 tokens

check-issues-status

Orchestrate check-issue-status across a list of issues to find which are already addressed in the code. Supports checking every open issue in a repository, issues authored by a specific user, or issues matching a custom search query, and these scopes combine. Use to triage a backlog for stale or already-implemented…

tomzx/agents · 126 tokens

apideck-unified-api

Apideck Unified API — one set of methods for 146+ SaaS connectors across CRM (21 connectors), Accounting (34), HRIS (58), ATS (11), File Storage (5), Issue Tracking (6), and Ecommerce (17). Route to Salesforce, HubSpot, QuickBooks, Xero, NetSuite, BambooHR, Workday, Greenhouse, SharePoint, Google Drive, Jira, Shopify…

apideck-libraries/api-skills · 167 tokens

implement

Implement a claimed backlog item as a focused, working change in the target repository's checkout.

Agent-Clubhouse/Goobers · 20 tokens

newticket

Plan and write one new backlog ticket from a short brief. Refreshes the checkout against the latest main, gathers project context (CLAUDE.md, existing tickets, the relevant area of the codebase), and writes a complete, well-formatted Markdown ticket to ztickets/open/ - .md following the convention in…

TweekFawkes/prompts · 169 tokens