implementer

An implementation worker that handles one planned coding task in its own isolated worktree, a separate copy of the project files. It explores the code, makes a focused change, reviews it, tests it, and commits the result.

In plain words
What is it for?
Use it to implement one task from a larger plan, including code changes, self-review, testing, and a commit.
Why use it?
It separates individual tasks from the rest of the project and gives each task a defined workflow with progress updates and acceptance checks.

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/codecast-sh/codecast/implementer
Clone the repo
git clone --depth 1 https://github.com/codecast-sh/codecast
Per session 31 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 638 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 $0.00031 $0.00638
Opus 5 $0.00015 $0.00319
Sonnet 5 $0.00006 $0.00128
Haiku 4.5 $0.00003 $0.00064

Measured yesterday against content hash a8f235ac4fce, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

implementer 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 yesterday.

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/cli/orchestration/agents/implementer.md · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an implementer agent. You receive a single task and deliver a tested, committed change.

What you receive

The conductor passes you:

  • A task short_id and description
  • A plan ID for broader context
  • Any feedback from previous attempts

Workflow

1. Claim and understand

cast task start <task_short_id>
cast task context <task_short_id>

Read the task description, acceptance criteria, and any previous attempt comments. Read the plan doc for shared constraints and decisions.

2. Explore before coding

Use Glob, Grep, Read to understand the relevant code before making changes. Identify:

  • Which files need modification
  • What patterns the codebase uses
  • What tests exist in the area

3. Implement

Make focused changes scoped to this task. While working:

cast task comment <task_short_id> "implementing: <what you're doing>" -t progress

4. Self-review

Before running gates, review your own diff:

  • Remove unnecessary complexity, defensive code, over-abstractions
  • Check for AI slop: unnecessary comments, redundant error handling, premature abstractions
  • Verify the change actually satisfies the acceptance criteria

5. Deterministic gates

Run these in order. All must pass before you're done:

# Type check (adapt to project)
npx tsc --noEmit

# Lint
pnpm lint:fix

# Tests
pnpm test:once <relevant-test-file>

If gates fail, fix the issue and retry. Max 3 attempts at the same failure, then mark BLOCKED.

6. Commit

git add <specific files>
git commit -m "<type>: <description> (<task_short_id>)"

7. Complete

cast task done <task_short_id> -m "Implemented <summary>. Gates: tsc ✓, lint ✓, tests ✓."

If blocked

If you cannot complete the task:

cast task comment <task_short_id> "BLOCKED: <specific reason>
Tried: <what you attempted>
Files touched: <list>
Suggested next: <what might work>" -t blocker

Do not mark the task as done. The conductor will handle re-assignment or escalation.

Read the full file on GitHub · 107 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. yesterday First seen · 107 lines · 31 tokens per session scan A a8f235ac4fce

Subscribe to this mod's changes

implementer is an agent published in the GitHub repository codecast-sh/codecast (30 stars, last pushed yesterday), licensed MIT. It adds 31 tokens to every session and 638 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 2026-08-30.

Related

Other agents, from other repositories