implement

implement is a skill for Claude Code from juicesharp/rpiv-mono. It costs 71 tokens per session (3,124 once invoked), scanned A, original, MIT.

A workflow for carrying out an approved technical plan stored in the project’s .rpiv/artifacts/plans/ directory. It applies the plan phase by phase and checks each phase against its stated success criteria.

In plain words
What is it for?
Executing a full phased plan or one selected phase, changing only the files assigned to that phase, and verifying the required results before continuing.
Why use it?
It prevents an agent from improvising beyond the agreed plan and makes progress easier to verify after each implementation phase.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

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/juicesharp/rpiv-mono/implement
Any agent
npx skills add juicesharp/rpiv-mono --skill implement
Clone the repo
git clone --depth 1 https://github.com/juicesharp/rpiv-mono

Made for: Claude Code.

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 implement

README.md
[![agentmods](https://agentmods.dev/badge/skills/juicesharp/rpiv-mono/implement.svg)](https://agentmods.dev/skills/juicesharp/rpiv-mono/implement)
Your own site
<a href="https://agentmods.dev/skills/juicesharp/rpiv-mono/implement"><img src="https://agentmods.dev/badge/skills/juicesharp/rpiv-mono/implement.svg" alt="Measured on agentmods" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,124 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 5.1 $0.00071 $0.03124
Opus 5 $0.00036 $0.01562
Sonnet 5 $0.00014 $0.00625
Haiku 4.5 $0.00007 $0.00312

Measured 2d ago against content hash 33df46d2b8ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

implement 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.

packages/rpiv-pi/skills/implement/SKILL.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Implement

You are tasked with implementing an approved technical plan from .rpiv/artifacts/plans/. These plans contain phases with specific changes and success criteria.

Input

$ARGUMENTS

The input above is <plan-path> [phase]:

  • First token is the plan path under .rpiv/artifacts/plans/.
  • Anything after it (e.g. "Phase 2") names a single phase to scope to.

Rules:

  • If a phase is named → single-phase (sequenced lane) mode. Phases run sequentially, in the plan's dependency order, so every earlier phase has already landed its files — your prerequisites are present in the working tree. Implement ONLY the named phase, touching only the files that phase owns. Hard rules for this mode:
    • Never implement, create, or edit another phase's files — not even if one is missing. You are one unit of a sequenced run; back-filling another phase's work duplicates it and corrupts the next lane.
    • A genuinely missing prerequisite is a hard error, not a decision. If a file your phase must read or edit (one an earlier phase owns) is absent, STOP and fail with prerequisite missing: <path> (expected from an earlier phase). Do not create it, do not ask_user_question about it, and do not defer your own edits — fail loudly so the run surfaces the ordering defect.
    • Never silently defer your phase's own edits. Apply every change the named phase owns, or fail. A phase that writes a partial slice and checks off nothing is worse than a clean failure.
    • Repo-located scratch lives under .rpiv/tmp/, nowhere else. Any file you create that is not in your phase's files: — a probe script, a fixture, a captured payload — goes under .rpiv/tmp/ (exempt from the workflow's scope floor) or outside the repo entirely. The floor counts untracked files too (git status -uall): a scratch file left at the repo root or in an undeclared directory is an undeclared write the downstream implement-scope-check flags. Delete repo-located scratch when you're done with it regardless.
    • Do not read, edit, or check off other phases' sections. Stop and print the closing block as soon as the named phase's own success criteria pass.
    • Record a #### Reconciliation directive for cross-phase test updates you can't apply yourself. Under build/vet's parallel implement lane, a correct change in YOUR phase can invalidate a test expectation that lives in a SIBLING phase's landed section — and single-phase mode forbids editing another phase's files. Do NOT edit the sibling's file and do NOT silently skip it: record a #### Reconciliation directive in your OWN phase's section. The directive is a machine-findable, machine-applicable find → replace against a single repo-root-relative target, so the downstream reconcile stage can apply it write-restricted to the plan's declared write-set:
      #### Reconciliation
      - `path/to/x.test.ts`: replace `expect(r).toBe(3)` → `expect(r).toBe(4)` — <one-line rationale>
      
      The find/replace spans may span multiple lines, but must carry no inner backticks — the spans are backtick-delimited, so a snippet containing a backtick (a template literal, markdown-in-a-test) is inexpressible in this form. For such content use the fenced form: the same - : replace — <rationale> header with NO inline spans, followed by a find: line + a fenced code block, then a replace: line + a fenced code block (indent the blocks under the list item; the fence line's own indentation is stripped from the captured content):
      #### Reconciliation
      - `path/to/x.test.ts`: replace — <one-line rationale>
        find:
        ```
        const s = `a ${b} c`;
        ```
        replace:
        ```
        const s = `a ${b} c d`;
        ```
      
      The target MUST be in the plan's declared write-set — the union of every phase's files:, twin-expanded (x.ts licenses x.test.ts) — the same authority the scope floor enforces; eligibility is plan-derived, never a filename convention, so any language's test or expectation file qualifies once declared. reconcile rejects (and reports) a directive against an undeclared path — route such an edit through validate's remediation arm or the owning phase. Record directives only for MECHANICAL expectation/consistency updates you can state concretely; a real logic change in a sibling's files is plan-level work (/skill:revise), not a reconciliation directive.

Read the full file on GitHub · 177 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. 2d ago Changed · +15 lines 33df46d2b8ba
  2. 6d ago First seen · 162 lines · 71 tokens per session scan A 0d6bb42226aa

Subscribe to this mod's changes

implement is a skill published in the GitHub repository juicesharp/rpiv-mono (750 stars, last pushed 3d ago), licensed MIT. It adds 71 tokens to every session and 3,124 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

recipe-create-meet-space

Create a Google Meet meeting space and share the join link.

googleworkspace/cli · 18 tokens

atmos-config

Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.

cloudposse/atmos · 31 tokens

workthreads

SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…

specstoryai/getspecstory · 126 tokens

story-readiness

Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…

Donchitos/Claude-Code-Game-Studios · 77 tokens

autotask-creator

Rules for automation CRUD from the group-chat commander. The commander does not call mutation tools and does not edit cloud/autotasks files directly. It emits one or more top-level ... containers in its final text; the bus parses and applies them after the turn.

Orkas-AI/Orkas · 5 tokens

monorepo-management

Master monorepo management with Turborepo, Nx, and pnpm workspaces to build efficient, scalable multi-package repositories with optimized builds and dependency management. Use when setting up monorepos, optimizing builds, or managing shared dependencies.

wshobson/agents · 54 tokens