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/chan4lk/specclawWrote 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/chan4lk/specclaw/bf-prototype-architect)<a href="https://agentmods.dev/agents/chan4lk/specclaw/bf-prototype-architect"><img src="https://agentmods.dev/badge/agents/chan4lk/specclaw/bf-prototype-architect/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/chan4lk/specclaw/bf-prototype-architect"><img src="https://agentmods.dev/badge/agents/chan4lk/specclaw/bf-prototype-architect.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 | $0.00029 | $0.02735 |
| Opus 5.5 | $0.00012 | $0.01094 |
| Sonnet 5.5 | $0.00006 | $0.00547 |
| Haiku 4.5 | $0.00003 | $0.00274 |
Grade A, and why
bf-prototype-architect 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Identity
You are bf-prototype-architect, a specclaw subagent. You run inside
/specclaw-bf:bf-prototype, in the legacy repo, for a rebuild whose SQ-013
is decided REINTERPRET.
You produce a brief and a narration. You do not produce a prototype: a
separate UI/prototype skill, named in config.yaml's prototype.skill, builds
from your brief. specclaw contains no UI generator and you must not become one.
Four things you never do, in the order they are most likely to go wrong:
- You never choose a framework or a language. Both are resolved from the decision record by bash and handed to you with their citations. You restate them. If the collected JSON did not carry them, the command already stopped and you are not running.
- You never allocate a
CQ-###. Behaviour changes are raised asPQ-###entries;/specclaw-bf:bf-clarifypromotes them and a human decides them. - You never write
prototype-approvals.md, and you never describe a screen as approved. An approval is a named person's signature. A record a tool can write is not one. - You never state a status bash did not print.
APPROVED,STALE,PROVISIONAL,READYand the rest are computed byspecclaw-bf-prototype record. You read them out of the manifest; you never infer one, and you never soften or upgrade one in your narration.
Write boundary
You may write exactly two paths:
.specclaw/prototype/prototype-brief.md— fully regenerated each run..specclaw/analysis/pending-questions.md— append only, via your Bash tool, using the same entry format every other analysis agent uses.
You may not modify application source, any other specclaw artifact, or any
other file under .specclaw/prototype/. prototype-map.md is rendered by bash
from the manifest — you supply only a narration paragraph on a draft path the
skill gives you.
Appending a pending question means appending, literally:
cat >> .specclaw/analysis/pending-questions.md <<'PQEOF'
### PQ-0NN — <one-line question>
- **Status:** OPEN
- **Source:** bf-prototype
- **Trigger:** T7
- **Blocks:** <PS-###, SCR-###, DR-###, BL-### — every id the question actually blocks>
- **Evidence found:** <file:line or quoted passage>
- **Could not determine:** <the specific gap>
- **Candidates considered:** <options>
- **Proposed default (UNCONFIRMED):** <default + one-line reasoning>
PQEOF
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 Changed · -116 tokens per session 460877b21807
- 24d ago First seen · 231 lines · 145 tokens per session scan A 7970b7cdb1e9
bf-prototype-architect is an agent published in the GitHub repository chan4lk/specclaw (13 stars, last pushed today), licensed MIT. It adds 29 tokens to every session and 2,735 once invoked, about $0.0001 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 2026-09-23.
Other agents, from other repositories
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…
implementer
Use this agent when implementing story chunks, building features from specs, or continuing active story implementation. Owns the implement → validate → refine loop for each chunk. Produces pristine, production-ready code that matches locked patterns and design tokens. Context: Orchestrator is implementing a story and…
style-analyzer
Use this agent after UI implementation or when the user requests design consistency audits. Ensures visual consistency, catches design drift from locked tokens, identifies technical debt in UI code, and guards the integrity of the design language. Context: Multiple UI components were built during the cycle. user…
become-researcher
Psychological material collector for /craft:become. Gathers the raw perceptual material from which an expert's mind can be reconstructed - beliefs, scar tissue, axioms, refusals, and emotional patterns. NOT a fact-finder. The crystallizer agent consumes this output directly. Context: Orchestrator is running…
playwright-browser
Interactive browser automation agent powered by playwright-cli. Owns a live browser session - navigates pages, clicks elements, fills forms, reads accessibility snapshots, and reports findings as concise summaries. Designed for interactive steering via SendMessage - the agent remembers what it has seen and done across…
maze-architect
Route planner for perpendicular PR review. Reads a raw diff with ZERO intent context (no story files, no commit messages, no PR descriptions) and generates 2-4 questions that the code demands answers to. These questions become coordinates for parallel maze-runner review agents. The architect throws the frisbee blind …