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 LuisUrrutia/skills --skill compare-solutionsgit clone --depth 1 https://github.com/LuisUrrutia/skillsWrote 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/luisurrutia/skills/compare-solutions)<a href="https://agentmods.dev/skills/luisurrutia/skills/compare-solutions"><img src="https://agentmods.dev/badge/skills/luisurrutia/skills/compare-solutions/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/luisurrutia/skills/compare-solutions"><img src="https://agentmods.dev/badge/skills/luisurrutia/skills/compare-solutions.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 | $0.00020 | $0.01421 |
| Opus 5.5 | $0.00008 | $0.00568 |
| Sonnet 5.5 | $0.00004 | $0.00284 |
| Haiku 4.5 | $0.00002 | $0.00142 |
Grade A, and why
compare-solutions 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 17d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compare solutions
Run independent complete attempts at one task, compare them, choose a base, incorporate useful parts, and verify the result. Use this when alternative approaches or independent investigations justify the extra work. Splitting a project into different subtasks is a separate coordination problem.
compare-solutions owns the comparison. A selected specialist owns how each
attempt does its work. For specialist candidates, read
references/specialists.md before framing the task.
Frame
Establish the deliverable, scope, source revision or input snapshot, acceptance requirements, and existing authority. Write a small task-specific rubric before seeing candidates: correctness, coverage, maintainability, or other observable criteria that distinguish a good result for this task. Give every candidate the same requirements and grounding; keep comparison notes and emerging preferences out of candidate prompts. A private rubric must not hide a requirement. Derive hard constraints from the caller's requirements and applicable rules; identify the source for each. Keep desirable implementation qualities as scored preferences. Neither the rubric nor the judge may turn a preference into a new reason for disqualification. Behavior outside the stated input domain can inform a documented tradeoff, but is not a contract failure without a requirement.
Use the host's delegation mechanism and available model catalog. Honor requested models, effort, candidate count, and budget. Otherwise start with two candidates; different available model families can broaden the comparison, but two independent contexts on one model remain useful. Report what actually ran. Do not guess model identifiers or silently replace an explicitly requested model.
Choose an output location for each candidate and for the synthesis. Follow the host's scratch, workspace, and ownership rules. Read-only candidates can share a stable source snapshot. Code candidates need separate mutable targets: isolated copies, patches against the snapshot, or workspaces provisioned through the authorized lifecycle. Separate filenames do not isolate a shared checkout, database, cache, or service. Arrange isolation before concurrent mutations; the coordinator alone writes the integrated target.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 17d ago Changed · -2 lines 7f1db033bf58
- 1mo ago First seen · 135 lines · 20 tokens per session scan A 9ca07127f207
compare-solutions is a skill published in the GitHub repository LuisUrrutia/skills (2 stars, last pushed today), licensed MIT. It adds 20 tokens to every session and 1,421 once invoked, about $0.0001 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 2026-10-06.
Other skills, from other repositories
capabilities-manager
Manage capa CLI configuration — capabilities.yaml / capabilities.json, skills, MCP servers, tools, hooks, sub-agents, rules, plugins, AGENTS.md / CLAUDE.md, security options, tool exposure modes, capa wrap shadow workspaces, --passthrough native writes, registries (including Claude marketplaces), activity traces, and…
bootstrap
Capify an existing project. Scans the repo for already-installed MCP servers, skills, rules, hooks, sub-agents, and plugins (across .claude, .cursor, .codex, .gemini, etc.), consolidates them into shared top-level directories (skills/, rules/, hooks/), and synthesizes a sound capabilities.yaml from the inventory. Use…
create-roadmap
Generates a battle-tested, phased master roadmap (progress.md) from any app idea or feature spec. Breaks development into foundational infrastructure, modular features, settings, monetization, security, polish, and store publishing, modeled after production systems like Floatkit.
project-scaffold
Scaffolds a complete project memory and multi-agent coordination system (gemini.md, AGENTSYNC.md, progress.md, taskplan.md, findings.md, crash-log.txt, .env.example, .gitignore). Provides a streamlined, stack-agnostic alternative to the B.L.A.S.T. protocol for deterministic development and seamless agent handoffs.
constraint-driven-development
Use when an Android project has no written quality bar, when the user says "set up constraints" or "define our standards", when coverage, performance, or accessibility should become enforced gates instead of per-PR arguments, when an agent keeps silencing checks or skipping tests to get to green, or when a threshold…
android-ui-engineering
Use when building UI with Jetpack Compose. Covers component design, state hoisting, recomposition optimization, Material 3 theming, Navigation 3, adaptive layouts for large screens, previews, screenshot verification, and XML interop.