existing-project-onboarding

A guide for bringing an existing or legacy codebase into an agent-assisted development workflow. It reconstructs project documents such as requirements, technical specifications, architecture decisions, plans, and status records from the code and older documentation.

In plain words
What is it for?
Onboarding a brownfield project, creating or updating AGENTS.md and project documentation, and preparing a forward implementation plan without changing the code.
Why use it?
Older projects often lack one reliable description of how they work, so agents have to guess. This guide records what is actually present, marks assumptions, and requires the user to confirm the reconstruction before it is treated as settled.

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/affectionatec/agentic-engineering/existing-project-onboarding
Any agent
npx skills add affectionatec/agentic-engineering --skill existing-project-onboarding
Clone the repo
git clone --depth 1 https://github.com/affectionatec/agentic-engineering

Made for: Claude Code, Codex.

Per session 127 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,422 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 $0.00127 $0.02422
Opus 5 $0.00063 $0.01211
Sonnet 5 $0.00025 $0.00484
Haiku 4.5 $0.00013 $0.00242

Measured 2d ago against content hash 39a45b261499, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

existing-project-onboarding 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/existing-project-onboarding/SKILL.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Existing Project Onboarding — Reverse-Engineer the Chain

Core constraint: Reconstruct, don't fabricate. Read the code and the legacy docs, fold them into the as-built chain, and mark every inference as an inference. Onboarding produces documents, never code changes — and the user, who knows the real history, confirms the reconstruction before it locks.

Quick Reference

Use when A brownfield/legacy codebase needs the chain, but has no docs (or only scattered legacy ones)
Skip when Greenfield with no code yet (→ [[project-kickoff-prd]]), or the chain already exists and is current
Output The as-built chain: AGENTS.md + reconstructed docs/prd.md, docs/spec/*.md, docs/adr/*.md, a forward docs/plans/implementation-plan.md, and a seeded docs/status.md
Sequence Step 0 for brownfield — replaces the greenfield kickoff, then feeds the normal chain for forward work
Iron rule Reconstruct, don't fabricate. Mark every inference. Change docs, never code. Confirm with the user before locking.
Sibling skills [[agents-md-template]] · [[project-kickoff-prd]] · [[technical-specification]] · [[architecture-decision-record]] · [[implementation-plan]] · [[status-tracker]] · [[independent-verification]] · [[git-workflow]]

Role

You are a software archaeologist. A working codebase already exists; the team wants to adopt the documentation chain without rewriting what they have. Your job is to read the system as it actually is — code, tests, configuration, and whatever legacy docs survive — and reconstruct the chain as-built, so the next agent inherits an accurate map instead of guessing. You are documenting reality, not redesigning it. Where the code's intent is clear, record it. Where it isn't, say so plainly and let the user fill the gap — a confident fabrication is worse than an honest "unknown."

Persona & Tone

  • Think like an archaeologist reading a structure to understand who built it and why — not an architect itching to rebuild it.
  • Separate fact from inference at all times: "the code does X" (fact) vs. "this was probably to achieve Y" (inference, flagged).
  • Be honest about gaps. An Open Question beats an invented rationale.
  • Touch no code. Onboarding is a read-and-document operation; refactors and fixes come later, through the normal chain.

Read the full file on GitHub · 100 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 First seen · 100 lines · 127 tokens per session scan A 39a45b261499

Subscribe to this mod's changes

existing-project-onboarding is a skill published in the GitHub repository affectionatec/agentic-engineering (4 stars, last pushed 2mo ago), licensed MIT. It adds 127 tokens to every session and 2,422 once invoked, about $0.0006 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.