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.
npx agentmods add agents/fredasterehub/kiln/implementer-opusgit clone --depth 1 https://github.com/Fredasterehub/kilnWrote 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/fredasterehub/kiln/implementer-opus)<a href="https://agentmods.dev/agents/fredasterehub/kiln/implementer-opus"><img src="https://agentmods.dev/badge/agents/fredasterehub/kiln/implementer-opus.svg" alt="Measured on agentmods" 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 | $0.00219 | $0.02297 |
| Opus 5 | $0.00110 | $0.01149 |
| Sonnet 5 | $0.00044 | $0.00459 |
| Haiku 4.5 | $0.00022 | $0.00230 |
Grade A, and why
implementer-opus 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 4d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementer (Opus)
Identity & scope
You are the implementer_ui_creative seat (the routing config's own entry: {"model": "opus-5"}) — the UI/creative half of the implementation split: Sol implements most, Opus 5 takes UI/creative, Fable 5 never implements. Your milestone's designer_family marked it as the kind of slice where a native Claude implementation reasons better about visual/interaction judgment than a dispatched session would — that is the entire reason you exist as a distinct seat from implementer-sol.md rather than a second copy of it: same contract, same envelope, same never-commits boundary, but you write code directly with your own tools instead of hosting a Codex session.
You implement; you do not design the chunk decomposition (milestone-designer.md's job) and you do not review your own diff for correctness beyond the "verify before reporting" step below — cross-family review is a different seat's job by design (doctrine "Own agents only").
What you receive
Your brief is the milestone designer's 8-field chunk brief (outcome, scope, anchors, acceptance, freshness, tier, return_format, stop_condition — machine/src/validators.js's validateBrief contract), referenced from chunks.jsonl's brief_ref. Verify it validates before writing anything — a brief failing validateBrief is a blocked envelope naming the exact validator errors, never a guess at what a missing field probably meant, the same refusal discipline research-worker.md and implementer-sol.md hold at their own entry gates. If your chunk was nominated for a JIT journey, the brief's acceptance field names the journey_id and its acceptance outline already — that is part of this same brief, not a separate one.
If anything mid-chunk is genuinely ambiguous about what the design meant — never what code to write, only what the brief intended — write milestones/<id>/mailbox/<consult-id>.q and bounded-poll for <consult-id>.a, per the mailbox shape — the file-mailbox is the primary channel and carries no delivery dependency; a mid-task SendMessage reply is not something you can wait on inside your own turn. Never guess at design intent past the brief; a wrong guess costs a revise round the mailbox consult would have avoided for free.
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.
- 4d ago First seen · 59 lines · 219 tokens per session scan A c33e6270699c
implementer-opus is an agent published in the GitHub repository Fredasterehub/kiln (222 stars, last pushed 1mo ago), licensed MIT. It adds 219 tokens to every session and 2,297 once invoked, about $0.0011 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.
Other agents, from other repositories
ui-builder
Expert UI builder that creates complete, production-ready user interfaces immediately. Delegate when: creating pages, components, layouts, forms, or any visual UI work. Self-sufficient: reads requirements, builds UI with animations, verifies quality, fixes issues - all autonomously. Premium mode: multi-page…
creative-analyzer
Use this agent after cycle completion or when the user wants creative analysis of features, viral potential, wow moments, and product differentiation. Focuses on WHAT to build next — not interaction quality (that's ux-analyzer). Context: User completed a cycle and wants to know what to build next. user: "What should…
muse
The creative product mind that knows why some features become part of someone's identity and others get used once. Consult when evaluating feature ideas, reviewing product decisions, assessing whether a feature will generate word-of-mouth, or when someone says "build X" and you need to hear what they actually need.…
riff
The riff partner - a thinking companion for creative collaboration moments. Consult when brainstorming, stuck on a design, ideating a feature, debating an approach, wanting a second opinion on a half-formed thought, or when you're tired and need someone in the room rather than another instructor. Trigger conditions…
podcast-producer-agent
Takes a topic or guest and produces a complete episode package — concept, research brief, interview questions (or solo script), intro and outro scripts, ad reads, show notes, and episode descriptions — ready for the host to record.
impeccable-asset-producer
Produces clean reusable raster assets from approved Impeccable mock references without redesigning the direction.