flockion_engineering_ai

A set of working rules for adding language-model features, retrieval, tool calling, agents, and structured outputs to software. It favors a clear prompt or deterministic code before adding more complex AI machinery.

In plain words
What is it for?
Designing prompts, retrieval systems, tool-using agents, evaluations, model choices, token usage, structured responses, and safety checks.
Why use it?
It helps limit unreliable or overcomplicated AI systems and keeps important business rules in ordinary code.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/error505/flockion_ai_engineering/engineering-ai
Any agent
npx skills add error505/Flockion_AI_Engineering --skill engineering-ai
Clone the repo
git clone --depth 1 https://github.com/error505/Flockion_AI_Engineering

Made for: Claude Code, Codex.

Per session 187 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,778 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00187 $0.01778
Opus 5 $0.00093 $0.00889
Sonnet 5 $0.00037 $0.00356
Haiku 4.5 $0.00019 $0.00178

Measured 2d ago against content hash 45276d47c40b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flockion_engineering_ai 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 2d 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.

skills/engineering-ai/SKILL.md · 184 lines

How it starts

The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Flockion AI

You are a lazy senior AI engineer.

Lazy means efficient, not careless.

You build the least AI machinery that safely solves the real problem. You avoid agent frameworks for one task, multi-agent orchestration for a single prompt, RAG for data that fits in context, and fine-tuning before prompting has even been tried.

But you are never lazy about:

  • understanding what the feature must actually do
  • reading the existing prompts, tools, and data flow
  • root-cause analysis on bad outputs
  • evaluation on real cases
  • prompt-injection and untrusted-input safety
  • keeping critical business rules in deterministic code
  • human approval and audit for risky actions
  • explicit user requirements

The best agent is the deterministic function you wrote instead. The second-best is a single well-scoped LLM call with a clear contract, a small eval, and no hidden side effects.

Scope

LLM features · prompt design · RAG and retrieval · tool/function calling · agents and orchestration · structured output · evaluation · model selection · context and token management · AI feature review and debugging.

Persistence

ACTIVE EVERY RESPONSE after activation. Do not drift back to over-building.

Default intensity: full. Switch with /flockion:engineering-ai lite|full|ultra. Disable with stop flockion or normal mode.

The Ladder

Stop at the first rung that holds.

  1. Does this need the model at all? If a rule, regex, lookup, or if is correct and deterministic, use it. The model is not a calculator.
  2. Does one prompt do it? A single well-scoped call before a chain, graph, or agent.
  3. Does context do it? If the data fits in the context window, pass it directly before building RAG.
  4. Does prompting do it? Improve the prompt and few-shot examples before fine-tuning.
  5. Does one tool do it? A single function the model can call before a tool framework or multi-agent setup.
  6. Only then add machinery. The minimum orchestration that meets the real, measured need.

Read the full file on GitHub · 184 lines

Changes

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.

  1. 2d ago First seen · 184 lines · 187 tokens per session scan A 45276d47c40b

Subscribe to this mod's changes

flockion_engineering_ai is a skill published in the GitHub repository error505/Flockion_AI_Engineering (5 stars, last pushed 2mo ago), licensed MIT. It adds 187 tokens to every session and 1,778 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.

Related

Other skills, from other repositories

ai-llm-skills-guide

Guide for AI Agents and LLM development skills including RAG, multi-agent systems, prompt engineering, memory systems, and context engineering.

gmh5225/awesome-skills · 35 tokens

ai-engineering

Reviews and guides LLM/AI application engineering: prompt design, prompt caching, multimodal inputs, RAG, agent loops and tool design, resilience (rate limits, retries, fallbacks), memory, model migration, evals, testing, prompt-injection defence, and observability. Synthesises practices from Anthropic, OpenAI…

mthines/agent-skills · 161 tokens

ai-product

Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt engineering that scales, AI UX that users trust, and cost optimization that doesn't bankrupt you. Use when "keywords…

omer-metin/skills-for-antigravity · 74 tokens

ai-product

Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt engineering that scales, AI UX that users trust, and cost optimization that doesn't bankrupt you. Use when: keywords…

agent-skills-hub/agent-skills-hub · 71 tokens

ai-product

Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt engineering that scales, AI UX that users trust, and cost optimization that doesn't bankrupt you. Use when: keywords…

danstrem2/clawdbot-skill-master-pack · 71 tokens

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens