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/frankaging/lite-research-agents/first-principlesnpx skills add frankaging/lite-research-agents --skill first-principlesgit clone --depth 1 https://github.com/frankaging/lite-research-agentsWrote 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/frankaging/lite-research-agents/first-principles)<a href="https://agentmods.dev/skills/frankaging/lite-research-agents/first-principles"><img src="https://agentmods.dev/badge/skills/frankaging/lite-research-agents/first-principles.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 | $0.00038 | $0.00880 |
| Opus 5 | $0.00019 | $0.00440 |
| Sonnet 5 | $0.00008 | $0.00176 |
| Haiku 4.5 | $0.00004 | $0.00088 |
Grade A, and why
first-principles 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- first-principles — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
First Principles
These principles are non-negotiable constraints on every research decision. Before proposing an idea, designing an experiment, or producing an artifact — check it against all five. If it fails one, rethink before proceeding.
The Principles
1. Simplicity is the primary quality signal. Always monitor the simplicity of the main artifact you are producing. Prefer the simpler solution. A good proxy: lines of code, number of components, number of moving parts. Do not stack ideas together. If a solution requires explaining multiple interacting novelties, it is probably too complex. Complexity is not depth.
2. Efficiency or performance — pick at least one, unambiguously. Work must move the needle on either: (a) efficiency without meaningful performance loss — faster inference, lower memory, fewer training FLOPs; or (b) raw performance. If a contribution does neither clearly, it is not ready. "Comparable performance with added complexity" is a failure mode.
3. Ask "what can someone actually do with this?" For every finding or method, ask: what is the concrete actionable insight for another researcher or practitioner? The best actionable insights satisfy principles 1 and 2. If the main takeaway is complex or does not improve efficiency or performance, the idea needs rethinking. If you cannot state the actionable insight in one sentence, it is not ready.
4. Humans stay in the loop. AI safety and human-AI collaboration are load-bearing assumptions, not caveats. Work that implicitly assumes full autonomy or that sidelines human oversight is out of scope. Every artifact should be legible and controllable by a human in the loop.
5. Do not work on problems that scale away. If a problem is likely to be solved by simply training larger models on more data, it is not a good research target. Before committing to a direction, ask: would a 10x scale-up plausibly dissolve this problem? If yes, redirect.
6. Do work on problems that scaling can newly enable. The inverse of principle 5 is an opportunity: find domains where scaling has not yet been applied, but would unlock something genuinely new if it were. Break the barrier — bring a scalable approach to a domain that has only seen small-scale work. The question is not "will scaling make this irrelevant?" but "could scaling make this possible for the first time?"
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 · 56 lines · 38 tokens per session scan A 5203b773fc54
first-principles is a skill published in the GitHub repository frankaging/lite-research-agents (3 stars, last pushed 5mo ago), licensed MIT. It adds 38 tokens to every session and 880 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 2026-08-31.
Other skills, from other repositories
meta-tags-optimizer
Optimize title tags, meta descriptions, Open Graph, and Twitter cards for maximum click-through rate. Generates multiple A/B test variations with character counting and SERP preview. Use when asked to "optimize title tag", "write meta description", "improve CTR", "Open Graph tags", "fix my meta tags", "social media…
google-ads-audit
Google Ads account audit and business context setup. Run this first — it gathers business information, analyzes account health, and saves context that all other ads skills reuse. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should…
paid-ads-linkedin
Audit, diagnose, plan, and safely operate connected LinkedIn Ads accounts through the NotFair MCP, with an export-based fallback. Use for LinkedIn advertising, sponsored content, lead-generation forms, job-title or company targeting, campaign groups, creatives, conversions, lead quality, budgets, bids, or approved…
sxo
Search Experience Optimization (SXO) — the bridge between SEO and UX/CRO. Audits the full journey from the SERP click to the on-page goal: SERP click-through factors (title/meta/rich results that win the click), then post-click experience signals that keep users and drive conversions — above-the-fold relevance and…
paid-ads-guide
Explain NotFair's paid-ads skills, installation, platform boundaries, account connections, and current product capabilities. Use for questions about how NotFair works, what it supports, how to install or connect it, plans or limits, or paid-media troubleshooting that is not an account-performance request.
arbor-agent-ideate
Strict IDEATE-stage skill for Arbor. Use immediately after TreeView(format="constraints") when drafting Idea Tree nodes, enforcing the ideadrafting and firstprinciplesprobe behavior, depth-aware idea levels, four-line TreeAddNode hypotheses, conflict checks, and self-filtering against shallow tweaks.