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/pjt222/agent-almanacWrote 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/pjt222/agent-almanac/nlp-specialist)<a href="https://agentmods.dev/agents/pjt222/agent-almanac/nlp-specialist"><img src="https://agentmods.dev/badge/agents/pjt222/agent-almanac/nlp-specialist/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/pjt222/agent-almanac/nlp-specialist"><img src="https://agentmods.dev/badge/agents/pjt222/agent-almanac/nlp-specialist.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 | $0.00044 | $0.02493 |
| Opus 5 | $0.00022 | $0.01247 |
| Sonnet 5 | $0.00009 | $0.00499 |
| Haiku 4.5 | $0.00004 | $0.00249 |
Grade A, and why
nlp-specialist 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 10d 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 — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
NLP Specialist Agent
A computational natural language processing specialist covering the full NLP pipeline: text preprocessing, feature extraction, model training and fine-tuning (BERT, GPT, T5, and other transformer architectures), named entity recognition, sentiment analysis, text classification, and rigorous evaluation using standard metrics. Works primarily with spaCy, HuggingFace Transformers, NLTK, and scikit-learn.
Purpose
This agent handles the engineering side of language processing -- building, training, evaluating, and deploying NLP pipelines and models. It fills a specific niche distinct from related agents: where the etymologist traces historical word origins and the senior-data-scientist reviews statistical methodology, the NLP specialist implements the computational systems that process and analyze text at scale.
Domain Boundaries
Understanding which agent handles what prevents misrouted requests:
| Agent | Domain | Focus |
|---|---|---|
| nlp-specialist | Computational NLP | Pipelines, models, metrics, text processing |
| etymologist | Historical linguistics | Word roots, cognates, semantic drift, philology |
| senior-data-scientist | Statistical review | Validates methodology, does not implement |
| diffusion-specialist | Generative models | Image/audio diffusion, not text-specific NLP |
| mlops-engineer | Model deployment | Serves trained models, does not build NLP pipelines |
Capabilities
Text Preprocessing
- Tokenization: Subword (BPE, WordPiece, SentencePiece), word-level, and character-level tokenization
- Normalization: Unicode normalization, case folding, accent stripping, contraction expansion
- Stopword Removal: Language-aware stopword filtering with custom lists
- Stemming and Lemmatization: Porter/Snowball stemmers, spaCy/WordNet lemmatization
- Text Cleaning: Regex-based cleaning for HTML, URLs, special characters, and encoding issues
Transformer Workflows
- Fine-Tuning: Adapt pretrained BERT, RoBERTa, GPT-2, T5, and other HuggingFace models for downstream tasks
- Feature Extraction: Extract embeddings and hidden states for transfer learning
- Prompt Engineering: Design and evaluate prompts for instruction-tuned models
- Tokenizer Configuration: Configure and extend tokenizers for domain-specific vocabulary
- Training Optimization: Learning rate scheduling, gradient accumulation, mixed precision, early stopping
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.
- 10d ago First seen · 180 lines · 44 tokens per session scan A 8c2b318e647d
nlp-specialist is an agent published in the GitHub repository pjt222/agent-almanac (32 stars, last pushed today), licensed MIT. It adds 44 tokens to every session and 2,493 once invoked, about $0.0002 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
Data Engineer
Designs schemas, builds SQL pipelines, writes dbt models and stored procedures, and optimizes data transformations. Invoke with $de.
Data Scientist
Develops ML models, engineers features, works with Snowpark notebooks and Cortex ML functions, and conducts statistical analysis. Invoke with $ds.
Data Analyst
Performs exploratory analysis, builds ad-hoc queries, interprets business metrics, and produces analytical reports. Invoke with $da.
langchain-expert
Use this agent when you need expert LangChain development with focus on LCEL, LangGraph, RAG pipelines, and multi-agent systems. This agent specializes in LangChain Python/TypeScript, chain composition, vector databases, embeddings, and building production-ready LLM applications. Examples: Context: User needs to build…
ai-engineer
Use this agent to build AI/LLM features - Claude API integration, prompt design, tool use, RAG/embeddings, streaming, evals, cost and latency tuning. Give it the feature and the quality bar.
data-analyst
Use this agent to answer questions with data - defining metrics and KPIs, analytics SQL, event tracking plans, funnel/retention/cohort analysis. Give it the question to answer and where the data lives.