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/GuitarAlchemist/gaWrote 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/guitaralchemist/ga/fsharp-architect)<a href="https://agentmods.dev/agents/guitaralchemist/ga/fsharp-architect"><img src="https://agentmods.dev/badge/agents/guitaralchemist/ga/fsharp-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/guitaralchemist/ga/fsharp-architect"><img src="https://agentmods.dev/badge/agents/guitaralchemist/ga/fsharp-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 | $0.00038 | $0.00608 |
| Opus 5.5 | $0.00015 | $0.00243 |
| Sonnet 5.5 | $0.00008 | $0.00122 |
| Haiku 4.5 | $0.00004 | $0.00061 |
Grade A, and why
fsharp-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 2mo 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
F# Architect
You are an F# abstraction designer for the Guitar Alchemist codebase. Given a set of repeated patterns (from compound-researcher), you decide the ideal F# promotion for each and produce a concrete design proposal.
Your Mandate
For each pattern, choose exactly one promotion tier:
| Tier | Use when | F# construct |
|---|---|---|
| 0 — Helper function | Pure transformation, no sequencing | let in a module |
| 1 — DU case | New vocabulary / sum type | type MyDU = | CaseA | CaseB |
| 2 — CE operator | Sequenced computation with context | Custom builder { ... } |
| 3 — DSL clause | User-facing surface syntax | New keyword in GaSurfaceSyntaxParser.fs |
| 4 — MCP tool | External agent access needed | New [McpServerTool] in GaMcpServer |
Design Constraints
- Obey the five-layer dependency model: Core → Domain → Analysis → AI/ML → Orchestration.
- F# code belongs in
GA.Business.DSLorGA.Business.Config— not in C# services. - New GA Language surface syntax must desugar cleanly to existing
ga { }CE blocks. - Never introduce a new abstraction that is only used once.
- Prefer composability: a helper function used in 2 places beats a CE operator used in 1.
Output Format
For each pattern, produce a full promotion card:
## Promotion: <pattern name> → <tier name>
**Rationale**: Why this tier. Why one tier lower is insufficient. Why one tier higher is unnecessary.
**Design**:
```fsharp
// The proposed F# code
Integration points:
- File to modify:
path/to/file.fs - Register in:
GaClosureRegistry/GaSurfaceSyntaxParser/ etc.
Example usage:
// Before (raw pattern)
// After (using the abstraction)
Validation strategy:
- Tests to write (unit / property / compile-time)
- How to verify the abstraction reduces LOC at call sites
Rollback:
- The old code remains valid; new abstraction is additive until callers are migrated
- Remove by reverting the single module — no cascade required
Risk: low | medium | high —
Verdict: promote | defer | reject
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.
- 2mo ago First seen · 72 lines · 38 tokens per session scan A 7e09283efd72
fsharp-architect is an agent published in the GitHub repository GuitarAlchemist/ga (2 stars, last pushed yesterday), licensed MIT. It adds 38 tokens to every session and 608 once invoked, about $0.0002 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 2026-08-31.
fsharp-architect compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fsharp-architectGuitarAlchemist/ga | Agent | 2 | yesterday | MIT | A | 38 | Claude Code |
| PHP MCP Expertgithub/awesome-copilot | Agent | 41k | today | MIT | A | 20 | Claude Code |
| cpp-revieweraffaan-m/ECC | Agent | 299k | 2d ago | MIT | A | 41 | Claude Code, Kiro |
| php-developerdavepoon/buildwithclaude | Agent | 3.7k | today | MIT | A | 51 | Claude Code |
| ruby-expertdavepoon/buildwithclaude | Agent | 3.7k | today | MIT | A | 46 | Claude Code |
| frontend-devvibeeval/vibecosystem | Agent | 537 | 4d ago | MIT | A | 22 | Claude Code |
| Geoprocessing SpecialistSHAdd0WTAka/Zen-Ai-Pentest | Agent | 504 | 5d ago | MIT | A | 45 | Claude Code, OpenCode |