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 skills/topprismdata/cultivating-ml-agent/knowledge-crystallization-feedback-loopnpx skills add topprismdata/cultivating-ml-agent --skill knowledge-crystallization-feedback-loopgit clone --depth 1 https://github.com/topprismdata/cultivating-ml-agentWrote 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/topprismdata/cultivating-ml-agent/knowledge-crystallization-feedback-loop)<a href="https://agentmods.dev/skills/topprismdata/cultivating-ml-agent/knowledge-crystallization-feedback-loop"><img src="https://agentmods.dev/badge/skills/topprismdata/cultivating-ml-agent/knowledge-crystallization-feedback-loop.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 5.1 | $0.00179 | $0.01917 |
| Opus 5 | $0.00089 | $0.00958 |
| Sonnet 5 | $0.00036 | $0.00383 |
| Haiku 4.5 | $0.00018 | $0.00192 |
Grade A, and why
knowledge-crystallization-feedback-loop 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 5d 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge Crystallization Feedback Loop
Problem
AI agents that compete in multiple ML competitions generate enormous experience, but without systematic crystallization:
- Same mistakes repeat (e.g., submitting without format check — 5 times)
- Hard-won insights are lost between sessions
- Knowledge files grow to 100+ items with no organization
- Retrieval fails when needed (can't find relevant lesson for new competition)
The Crystallization Cycle
┌─────────────────────────────────────────┐
│ 1. EXPERIMENT │
│ Run competition / try approach │
└──────────────┬──────────────────────────┘
▼
┌─────────────────────────────────────────┐
│ 2. IDENTIFY OUTCOME │
│ Success? Failure? Marginal? Dead end? │
└──────────────┬──────────────────────────┘
▼
┌─────────────────────────────────────────┐
│ 3. EXTRACT PATTERN │
│ What's the 1-sentence rule? │
│ What evidence supports it? │
│ When does it apply? When NOT? │
└──────────────┬──────────────────────────┘
▼
┌─────────────────────────────────────────┐
│ 4. CLASSIFY │
│ feedback = "don't do X" (anti-pattern)│
│ learned = "do X for result Y" │
│ reference = "X is located at Y" │
└──────────────┬──────────────────────────┘
▼
┌─────────────────────────────────────────┐
│ 5. STORE WITH TRIGGERS │
│ Frontmatter description = when to use │
│ Tags = search keywords │
│ Links = [[related-skills]] │
└──────────────┬──────────────────────────┘
▼
┌─────────────────────────────────────────┐
│ 6. ACTIVATE ON MATCH │
│ New task matches trigger → load skill │
│ Verify still current → apply or update│
└─────────────────────────────────────────┘
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.
- 5d ago First seen · 181 lines · 179 tokens per session scan A b6a9900d4ade
knowledge-crystallization-feedback-loop is a skill published in the GitHub repository topprismdata/cultivating-ml-agent (4 stars, last pushed 8d ago), licensed MIT. It adds 179 tokens to every session and 1,917 once invoked, about $0.0009 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 2026-08-31.
Other skills, from other repositories
huggingface-hub
Hugging Face Hub CLI (hf) — search, download, and upload models and datasets, manage repos, query datasets with SQL, deploy inference endpoints, manage Spaces and buckets.
tensorboard
Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit.
mlflow
Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow - framework-agnostic ML lifecycle platform.
datachain-knowledge
Use whenever datasets, cloud storage buckets, or data pipelines are mentioned — creating, saving, querying, listing, exploring, deleting, or processing data in S3, GCS, Azure Blob, or local storage. Also use when running any script that may create datasets as a side effect. Maintains a knowledge base at dc-knowledge/…
adding-models
Guide for adding new LLM models to Letta Code. Use when the user wants to add support for a new model, needs to know valid model handles, or wants to update model-specific compatibility behavior. Covers runtime catalog sources, CI test matrices, and handle validation.
prompt-scanner
使用 agent-sec-cli 扫描 prompt 文本中的注入攻击和越狱尝试，返回结构化 JSON 扫描结果。当用户提到 prompt 安全、prompt 注入检测、越狱检测、提示词攻击检测，或者需要判断一段文本是否包含恶意 prompt 注入时，都应使用此技能。即使用户没有明确说"扫描"，只要涉及评估 prompt 文本的安全性，也应触发此技能。.