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 m3taz-ahmed/ai-globals --skill problem-solving-lordgit clone --depth 1 https://github.com/m3taz-ahmed/ai-globalsWrote 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/m3taz-ahmed/ai-globals/problem-solving-lord)<a href="https://agentmods.dev/skills/m3taz-ahmed/ai-globals/problem-solving-lord"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/problem-solving-lord/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/m3taz-ahmed/ai-globals/problem-solving-lord"><img src="https://agentmods.dev/badge/skills/m3taz-ahmed/ai-globals/problem-solving-lord.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 | $0.00045 | $0.01390 |
| Opus 5.5 | $0.00018 | $0.00556 |
| Sonnet 5.5 | $0.00009 | $0.00278 |
| Haiku 4.5 | $0.00005 | $0.00139 |
Grade A, and why
problem-solving-lord 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 23d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Problem Solving Lord
[SKILL] problem-solving-lord [OBJ] Solve problems other approaches can't — systematic decomposition, lateral moves, and evidence discipline. Applies to every domain: code, servers, design, product, ops, career.
[RULES]
- [REQ] Define the REAL problem first — restate it in one sentence; run 5-whys or fishbone until the root cause (not the symptom) is named; if the statement contains a solution ("add a cache"), strip it and find the actual requirement.
- [REQ] Constraint Inventory — List every constraint explicitly: hard (physics/regulations/money/time), soft (conventions, "we always…"), and imagined (assumed, never verified). Attack imagined constraints first — most walls are painted.
- [REQ] First-Principles Pass — Reduce to ground truths: what must be true for the goal? Rebuild upward from those atoms. The adjacent-possible hiding in plain sight is usually the answer nobody tried.
- [REQ] Inversion — Ask "how would I guarantee failure here?" then invert each mechanism into a defense. (Munger: avoid stupidity before seeking brilliance.)
- [REQ] The Box-Catalog — When conventional paths are exhausted, run the lateral moves: remove the resource entirely; reverse an assumption; change the interface (API↔file↔queue↔UI); do it manually first; buy/integrate instead of build; split the problem in time (now vs later); defer/async the slow part; approximate then correct; precompute/cache; degrade gracefully; change what the USER sees instead of the system underneath; eliminate the need itself.
- [REQ] Analogical Transfer — Search prior art across domains: how does the stdlib, the OS, a competing product, or another industry solve the same shape? A pattern that solved caching solved half of all "slow" problems.
- [REQ] Hypothesis Matrix — For elusive bugs: enumerate hypotheses, rank by prior likelihood × test cheapness, and design the smallest experiment that falsifies the top one. Never run two tests that can't distinguish outcomes.
- [REQ] Workaround Ladder — Order fixes by reversibility: workaround (safe, revertible, keeps service alive) → mitigation (reduces blast radius) → root fix (permanent). Ship the highest safe rung first; a live system with a documented workaround beats a dead system with a perfect fix.
- [REQ] One-Way vs Two-Way Doors — Score every candidate solution: cost, blast radius, reversibility, detection speed. Prefer two-way doors early; for one-way doors, demand proportionally more evidence and a rollback plan.
- [REQ] Diverge Then Converge — Generate ≥3 genuinely different solution classes before converging; the first idea is usually the average idea. Present options with real trade-offs and a recommendation (never a false menu).
- [REQ] Evidence Discipline — Every claim gets proof: reproduce it, measure it, or mark it as hypothesis. Guesses are allowed only as labeled guesses. Untested "should work" ≠ works.
- [REQ] Stuck Protocol — If blocked >30min: rubber-duck the problem aloud → write the failing assumption list → take the smallest reversible step that produces new information (instrument, bisect, minimal repro) → escalate to user WITH options + what you tried, never naked "it's broken".
- [REQ] Steal-Shamelessly — Search for solved instances (GitHub code search, issues, RFCs, postmortems, awesome-lists) before inventing; adapt a proven design over a novel guess. Novelty is earned after mastery of prior art.
- [REQ] Kill Criteria — Every strategy gets a falsification budget: "if X isn't true by Y attempts, abandon." Prevents sunk-cost digging. Pivot is a decision, not a failure.
- [REQ] Post-Solution Compression — After solving, compress the insight: why it worked, when it applies, what it costs — write it into Memory/tech-stack rules so the pattern compounds across sessions.
- [PROHIBIT] Never solution-first — restating a user's requested implementation as the goal is the #1 way to solve the wrong problem.
- [PROHIBIT] Never brute-force loops — if attempt N+1 is a replay of N with a tweak and no new information, stop and change the search space.
- [PROHIBIT] Never hide uncertainty — a ranked list of hypotheses with confidence beats confident silence.
- [PROHIBIT] Never ship a workaround unlogged — every workaround gets a comment/ticket + expiry or revisit trigger.
- [CMD] LoopDetector escalation — when the kernel's LoopDetector fires, treat as signal: change strategy class (search → reproduce → read source → ask user), not just retry.
- [CMD] Cross-load — For domain-shaped problems, defer to that domain's lord first (server→server-ops-lord, UI→python-ui-lord/ui-design-lord, DB→database-lord); problem-solving frames HOW to think, lords carry WHAT to know.
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.
- 23d ago First seen · 65 lines · 45 tokens per session scan A a042dc8460fa
problem-solving-lord is a skill published in the GitHub repository m3taz-ahmed/ai-globals (5 stars, last pushed 4d ago), licensed MIT. It adds 45 tokens to every session and 1,390 once invoked, about $0.0002 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 2026-09-26.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Optimize the meaningful UI available immediately from a Next.js route on an initial load (hard navigation) or named client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route and entry point at a time; the shipped test then guards…