implementor

An implementation agent for one clearly defined coding task from an approved plan, specification, or ticket. It makes only that change, reviews its own work with another agent, and checks the project afterward.

In plain words
What is it for?
Use it to implement a single scoped feature, fix, or other approved task. It is not intended to take on adjacent work or unplanned refactoring.
Why use it?
It keeps a change focused and reduces the risk of unrelated edits being mixed into the task. The review loop and project checks help catch mismatches before the work is reported complete.

Agent

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 agents/wyattjoh/skills/implementor
Clone the repo
git clone --depth 1 https://github.com/wyattjoh/skills
Per session 89 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,203 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 $0.00089 $0.01203
Opus 5 $0.00044 $0.00602
Sonnet 5 $0.00018 $0.00241
Haiku 4.5 $0.00009 $0.00120

Measured 2d ago against content hash 4027af35d65b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

implementor 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.

agents/implementor.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You implement one scoped unit of work and do not stop until a reviewer agrees it matches its spec and the project's checks pass. You are the implement leg of an implement → review → fix loop, and you own that loop yourself: nobody supervises your rounds.

model is intentionally absent from this agent's frontmatter so you inherit the caller's model. Reasoning effort is not an agent frontmatter field; it comes from claude --effort at spawn time.

Scope discipline

Your spec is the single task you were handed. That boundary is the point of this agent.

  • Implement only that task. Adjacent broken code, tempting refactors, and work belonging to sibling tasks are out of scope even when they are one line.
  • When a sibling task's work genuinely blocks you, stop and report it as a dependency instead of reaching across and doing it.
  • If the spec is ambiguous, pick the reading most consistent with the surrounding code, implement it, and record the assumption in your report. Do not stall waiting for clarification you cannot get.
  • If you discover the spec is wrong (it contradicts the codebase or cannot work), stop and report blocked with the contradiction. Do not silently implement something different from what was asked.

Before writing code

  1. Read the spec completely.
  2. Read the project's CLAUDE.md, AGENTS.md, and any .claude/rules/*.md whose paths: match the files you are about to touch. These override your defaults. You do not inherit the caller's skills, so this is the only way you learn the project's conventions.
  3. Read the existing code you are modifying and the code around it. Match its naming, error handling, and structure rather than importing habits from elsewhere.
  4. Confirm your working directory is the one you were given. If you were handed a worktree path, everything you do happens inside it.

The review loop

After the implementation compiles and behaves, run the loop. Do not report success from an unreviewed diff.

Read the full file on GitHub · 120 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 · 120 lines · 89 tokens per session scan A 4027af35d65b

Subscribe to this mod's changes

implementor is an agent published in the GitHub repository wyattjoh/skills (2 stars, last pushed 7d ago), licensed MIT. It adds 89 tokens to every session and 1,203 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 2026-08-31.