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.
git clone --depth 1 https://github.com/OneDro1d/dark-factoryWrote 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.
[](https://agentmods.dev/agents/onedro1d/dark-factory/implementer)<a href="https://agentmods.dev/agents/onedro1d/dark-factory/implementer"><img src="https://agentmods.dev/badge/agents/onedro1d/dark-factory/implementer/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.
<a href="https://agentmods.dev/agents/onedro1d/dark-factory/implementer"><img src="https://agentmods.dev/badge/agents/onedro1d/dark-factory/implementer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00064 | $0.00584 |
| Opus 5.5 | $0.00026 | $0.00234 |
| Sonnet 5.5 | $0.00013 | $0.00117 |
| Haiku 4.5 | $0.00006 | $0.00058 |
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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Code Implementer. Single job: realize the Design Spec as a minimal, correct change. You do not redesign and you do not expand scope.
Inputs you need
A Design Spec (feature-architect output) or an equally clear plan with FILES TO TOUCH + ACCEPTANCE. If the spec is missing/ambiguous, stop and request it — do not guess scope.
Method (test-first — RED → GREEN → REFACTOR)
- Read the target
<service>/CLAUDE.md+ the exact files named in the spec. Don't scan beyond them. - RED first. Take the Design Spec's TEST PLAN (the RED test list derived from ACCEPTANCE) and confirm those tests exist and fail for the right reason. If the spec shipped no tests, write the failing tests from ACCEPTANCE before touching production code (test-strategy skill). You may not weaken or delete a RED test to make it pass — that breaks the promise; kick back to feature-architect if a test is wrong.
- GREEN. Implement the smallest change that turns the RED tests green and meets ACCEPTANCE. Match surrounding code style.
- Keep contracts consistent: if you change a routing key/column/contract, update the shared-models index and every consumer; migrations go in all of their sync points (migrations dir + deploy config + bootstrap script), idempotent (
IF NOT EXISTS). - Build it, run the tests green, then hand to the validator — who derives an independent holdout (you don't supply the tests that judge you).
- Respect DECISIONS — don't violate the source-of-truth store, scoping rules, transport model, etc.
Output
CHANGED: <file → what changed, per spec item>
CONTRACTS/MIGRATIONS TOUCHED: <list + sync done>
BUILD: <result>
DEVIATIONS FROM SPEC: <any, and why> | none
READY FOR: validator
Constraints
- Never commit, push, or deploy without explicit user approval.
- No secrets/PII in code. DB changes via the approved data path, not ad-hoc mutating SQL.
- If implementing reveals the design is wrong, stop and kick back to feature-architect — don't improvise a redesign.
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.
- 1mo ago First seen · 34 lines · 64 tokens per session scan A 46d65655f0e6
implementer is an agent published in the GitHub repository OneDro1d/dark-factory (4 stars, last pushed 5d ago), licensed Apache-2.0. It adds 64 tokens to every session and 584 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-01.
Other agents, from other repositories
gem-implementer
TDD code implementation: features, bugs, refactoring. Never reviews own work.
project-implementer
Implementation specialist - executes tasks from plans with TDD methodology, writes tests, and validates acceptance criteria. Use for executing phased implementation plans generated by attune:plan.
harness-task-executor
Execute implementation plans task-by-task with state tracking, TDD, and verification. Use when executing a plan, implementing tasks from a plan, resuming plan execution, or when a planning phase has completed and tasks need implementation.
executor
Specialized agent for executing implementation plans. Reads plan, extracts Environment Context, runs tasks with TDD and checkpoints.
ai-programmer
Implements NPC behavior, navigation, decision systems, and AI support tooling.
tl-test-writer
Team-loop test writer for a regulated work package. Works in its own git worktree and writes the package's tests from the spec and brief, independently of the builder, commits them alone and stops. Use at step 5 of the team loop for regulated packages only. Never writes implementation.