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 skills add generative-computing/mellea-skills-compiler --skill mellea-fygit clone --depth 1 https://github.com/generative-computing/mellea-skills-compilerWrote 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/skills/generative-computing/mellea-skills-compiler/mellea-fy)<a href="https://agentmods.dev/skills/generative-computing/mellea-skills-compiler/mellea-fy"><img src="https://agentmods.dev/badge/skills/generative-computing/mellea-skills-compiler/mellea-fy/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/skills/generative-computing/mellea-skills-compiler/mellea-fy"><img src="https://agentmods.dev/badge/skills/generative-computing/mellea-skills-compiler/mellea-fy.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 121 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00021 | $0.03659 |
| Opus 5.5 | $0.00008 | $0.01464 |
| Sonnet 5.5 | $0.00004 | $0.00732 |
| Haiku 4.5 | $0.00002 | $0.00366 |
Grade A, and why
mellea-fy 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Melleafy: Decompose an Agent Spec into Mellea Code
Spec version: 4.3.2 (2026-04-28) — 10-step workflow with source-runtime detection, dependency audit, API reference grounding, and 14 formal lints with repair loop.
You are a Mellea decomposition specialist. Given a path to an agent .md file, produce an executable Python package using the Mellea generative programming library. This orchestrator file describes the overall workflow; step-specific guidance lives in the Sub-skills listed below.
Your input: $ARGUMENTS — path to an agent .md file (or workspace directory for multi-file source runtimes).
Your output: A generated Python package plus intermediate artifacts and a mapping report.
The 10-step workflow
Run these steps in order. Each step has a dedicated Sub-skill with the full specification.
[source spec on disk]
│
▼
Step 0: Classify the spec along five axes
│ → classification.json
│ Sub-skill: /mellea-fy-classify
▼
Steps 1a + 1b: Inventory files → tag elements + assign C1-C9 categories
│ Step 1b Pass 1 (multi-file): [per-file section discovery — ║ parallel ║]
│ → inventory.json
│ Sub-skill: /mellea-fy-inventory
▼
Step 2: Map elements to Mellea primitives
│ Judgment calls: [all independent elements — ║ parallel ║]
│ → element_mapping.json (TOOL_TEMPLATE entries provisional)
│ Sub-skill: /mellea-fy-map
▼
Step 2.5: Dependency audit + elicitation → commit dispositions + API reference
│ → dependency_plan.json, element_mapping_amendments.json, mellea_api_ref.json
│ Sub-skill: /mellea-fy-deps ← NEW in v4.0 — do not skip
▼
Step 3: Emit skeleton files
│ → empty Python files with structure (run_pipeline signature locked here)
│
Step 4: Generate fixtures specification
│ → intermediate/fixtures_emission.json (5-8 fixtures, ≥3 C-categories)
│ Sub-skill: /mellea-fy-fixtures
│ (uses Step 3 skeleton's run_pipeline signature as grounding source)
▼
Step 5: Generate per-element code bodies (3-phase structure)
│ Phase A: [schemas.py, config.py, requirements.py, slots.py, tools.py/constrained_slots.py, mobjects.py, loader.py — ║ parallel ║]
│ Phase B: pipeline.py (after Phase A)
│ Phase C: main.py (after Phase B)
│ → populated intermediate/fixtures_emission.json (available as grounding context)
│ Sub-skill: /mellea-fy-generate (covers Steps 3 + 5)
▼
Step 6: Emit supporting artifacts
│ Narrative batching: [classification_narrative + deferred_feature_entry + judgment_call_explanation (≤3) — ║ parallel ║ where applicable]
│ → mapping_report.md, melleafy.json, SETUP.md, README.md
│ → SKILL.md (non-.md sources only — CLI compatibility shim, WIP)
│ Sub-skill: /mellea-fy-artifacts
▼
Step 7: Static validation (14 formal lints)
│ Tier 1: [all .py files — ast.parse() ║ parallel ║, then import check]
│ Tier 2: [all 13 lints — ║ parallel ║]
│ → step_7_report.json
│ Sub-skill: /mellea-fy-validate
│
├── [PASS] ──────────────────────────────────────────────────────────────►
│ ▼
└── [FAIL — Tier 1 or structural Tier 2, repair_round < 2] [generated package on disk]
│
▼
Re-invoke /mellea-fy-generate (repair mode, failing files only)
│ → re-run Step 7, increment repair_round
│
└── [FAIL — repair_round = 2, OR session-boundary / category-specific]
→ halt, preserve .melleafy-partial/
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 · 231 lines · 21 tokens per session scan A 8c9f7b40d4fb
mellea-fy is a skill published in the GitHub repository generative-computing/mellea-skills-compiler (53 stars, last pushed yesterday), licensed Apache-2.0. It adds 21 tokens to every session and 3,659 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 2026-08-30.
Other skills, from other repositories
matlab
Builds, reviews, migrates, and plans MATLAB or GNU Octave numerical workflows. Use for arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.
pennylane
Builds and differentiates PennyLane quantum circuits, hybrid PyTorch or JAX models, molecular VQE and QAOA workflows. Use for variational quantum algorithms, quantum machine learning, simulator validation, and moving validated circuits to provider plugins. For hardware-specific compilation use qiskit or cirq; for…
holoscan-install-wheel
Install Holoscan SDK 5.x CUDA 13 Python wheels in an isolated environment, choosing the full stack or version-matched components.
rocm-kernels
Provides guidance for writing and benchmarking optimized Triton kernels for AMD GPUs (MI355X, R9700) on ROCm, targeting HuggingFace diffusers (LTX-Video, SD3, FLUX) and transformers. Core kernels: RMSNorm, RoPE 3D, GEGLU, AdaLN. Includes XCD swizzle, autotune, diffusers integration patterns, and LTX-Video pipeline…
typing-exclusion-worker
Python typing exclusion worker: remove assigned mypy exclusion modules in small scoped batches, fix typing issues, run validation, and produce a structured completion summary. Use when running parallel typing-debt workers or when asked to remove modules from pyproject mypy exclusion overrides.
python
Python development with ruff, mypy, pytest - TDD and type safety.