A one-time migration procedure for bringing an existing project into the Workbench while preserving its code, history, project information, and remote recovery point.

Use it to inventory and migrate an existing checkout, verify its remote and controls, archive retired documents, and commit and push the migration work.

It provides checks for dirty or diverged Git state and prevents the migration from silently losing work, rewriting history, or changing remote access assumptions.

adoption — Skill for Claude Code and Codex by KaydenClark

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo

git clone --depth 1 https://github.com/KaydenClark/LLM_Workbench

agentmods

npx agentmods add skills/kaydenclark/llm_workbench/adoption

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/adoption/github.svg)](https://agentmods.dev/skills/kaydenclark/llm_workbench/adoption)

Your own site

<a href="https://agentmods.dev/skills/kaydenclark/llm_workbench/adoption"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/adoption/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/kaydenclark/llm_workbench/adoption"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/adoption.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 836 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.00028	$0.00836
Opus 5	$0.00014	$0.00418
Sonnet 5	$0.00006	$0.00167
Haiku 4.5	$0.00003	$0.00084

Measured today against content hash 738f133d3239, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

adoption 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 today.

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/adoption/SKILL.md · 65 lines

What it actually says

Use Adoption only when an existing project is joining the Workbench for the first time. An already-adopted room that still sits on a v2 root moves onto the v3 support root through /update-harness, which runs node tools/workbench-upgrade.mjs upgrade --layout-only and records provenance.lifecycle: upgrade; do not rerun Adoption for it, because a second adoption record contradicts its first. A routine harness update also uses /update-harness.

Verify the canonical checkout, branch, dirty/ahead/diverged state, remote, and nearest controls. Dirty state routes recovery: checkpoint owned work or isolate the migration; never abandon the run merely because the tree is not clean.
Locate the canonical Workbench source named by the workspace controls and read templates/ADOPTION.md completely. Follow its one-time inventory, provenance, migration, verification, and handoff phases. After reconciling the seven filled root controls, install or verify the core bundle in the intended disposable or user-scoped home and run:
```
node tools/workbench-adoption.mjs migrate \
  --project [ABSOLUTE_PROJECT_PATH] \
  --home [USER_HOME] \
  --version v3.2.0
```
The helper moves only unambiguous durable v2 lanes, including a legacy root MEMORY.md and a root feedback file, into the manifest-declared schema 2 lanes and collections, preserves a project-local skills/ folder under workbench/sessions/recovery/, installs the receipt-backed runtime tools into workbench/tools/, declares git.integrationBranch by the exact case of an existing integration-named branch (listing an unresolved one as residue.missingIntegrationBranch), renders the projections, and runs doctor. An existing support root, path collision, missing core skill, or bracketed root control is a blocker to reconcile before retrying; never overwrite it, and never touch an application's root tools/. Afterwards use the project's own node workbench/tools/spec-workbench.mjs ... copies.
Preserve project behavior and history. Port live truth into the existing Workbench owners, archive retired control documents, and do not perform an unrelated product cleanup or create another tracker.
Verify remote recovery before broad edits. When current authorization and an authenticated namespace permit it, create a private remote if the canonical project has none. Never infer public visibility, rewrite remote history, change credentials, or replace an existing remote.
Always commit and push each coherent migration ticket and every incomplete checkpoint on a prefixed task branch; untracked migration output is not a finished Adoption. Promote verified work only to the declared integration branch (git.integrationBranch in workbench/manifest.json; the migration declares an existing integration-named branch or integration). When authorization permits, create it from the default branch and push it; otherwise record the omission reason in the owning spec. The merge from the declared branch into the default branch stays with the owner.
Record the source remote, ref, resolved commit, executed self-tests, and any vendored-helper checksum in the owning spec. Put fresh-clone reproduction commands in RUNBOOK.md, validate workbench/manifest.json, resolve the next ticket from its declared workbench/specs/ lane, and report the pushed recovery ref and remaining owner gates. Append observed harness friction to the manifest-declared feedback lane; if none was observed, record none observed with the reason in the owning spec.

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.

today Changed · -1 lines 738f133d3239
yesterday Changed c57849136029
3d ago Changed · +34 lines 9e39c6cf3f4f
5d ago First seen · 32 lines · 28 tokens per session scan A 16954037cde1

Subscribe to this mod's changes

adoption is a skill published in the GitHub repository KaydenClark/LLM_Workbench (2 stars, last pushed today), licensed MIT. It adds 28 tokens to every session and 836 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 2026-09-05.

Git and workflow

prowler-commit

Creates professional git commits following conventional-commits format. Trigger: When creating commits, after completing code changes, when user asks to commit.

prowler-cloud/prowler · 33 tokens

gh-auth-isolation

Safely manage multiple GitHub identities (EMU + personal) in agent workflows.

github/gh-aw · 20 tokens

comet-github

A routing guide for Comet-related GitHub work. It directs requests about pull requests, issues, CI failures, ideas, and fixes to the appropriate review or implementation process.

rpamis/comet · 72 tokens

github-skill

Work with GitHub via the gh CLI — clone repositories, create/list/merge pull requests, create/list issues, and run any other gh command (API calls, workflow runs, releases, repo administration). List operations return parsed JSON.

zeenie-ai/OpenCompany · 51 tokens

re0-merge

Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…

LilMGenius/paperthin · 70 tokens

nvca-chart-release

Release NVCA Operator chart changes from the native monorepo source to the vendored Helm chart. Use when updating the vendored NVCA Operator chart, changing NVCA image refs, publishing helm-nvca-operator, or validating the chart against a self-managed control plane.

NVIDIA/nvcf · 61 tokens