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 bikeread/promethos --skill evolve-skill-librarygit clone --depth 1 https://github.com/bikeread/promethosWrote 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/bikeread/promethos/evolve-skill-library)<a href="https://agentmods.dev/skills/bikeread/promethos/evolve-skill-library"><img src="https://agentmods.dev/badge/skills/bikeread/promethos/evolve-skill-library/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/bikeread/promethos/evolve-skill-library"><img src="https://agentmods.dev/badge/skills/bikeread/promethos/evolve-skill-library.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 | $0.00029 | $0.00701 |
| Opus 5.5 | $0.00012 | $0.00280 |
| Sonnet 5.5 | $0.00006 | $0.00140 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
evolve-skill-library 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 2mo 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Goal
Keep the skill library coherent by merging overlaps, retiring stale material, and promoting only the patterns that earn long-term reuse.
Inputs
- Current skill inventory
- Usage history or anecdotal friction
- Repeated failures, corrections, or missing patterns
Non-Goals
- Rewriting the entire library from scratch without evidence
- Expanding scope just to make the library feel more complete
Workflow
Trigger signals
- Two skills seem to overlap in scope
- A skill has not been triggered in a long time
- Users complain the library is confusing or hard to navigate
- Skill count is growing without clear value
1. Inventory the library by behavior, not by file count
List each skill's purpose, trigger, current usefulness, and nearest neighbors. Note which skills have strong evidence of repeated use and which only look important in theory. Success criteria: The library is mapped as a set of decision points, not a flat list of filenames.
2. Detect overlap, gaps, and drift
Compare neighboring skills to find duplicate triggers, missing coverage, stale assumptions, and places where a skill has started to sprawl beyond its original decision area. Success criteria: Each suspicious area is labeled as overlap, gap, or drift, so the next action is obvious.
3. Check for buried or drifting guardrails
Review whether risky behavior, approval logic, or escalation rules have drifted out of the guard layer and into general-purpose skills. Success criteria: Cross-cutting guardrail logic is either kept visible in the guard layer or intentionally referenced there.
4. Decide keep, merge, split, archive, or promote
Apply one explicit rule set to each candidate change: preserve a skill that is still distinct and useful, merge only when two skills answer the same decision, split only when one skill contains multiple decision points, archive when the pattern is no longer in use, and promote when a repeated rule has earned a skill. Success criteria: Every decision has a rationale tied to usage evidence or library clarity.
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.
- 2mo ago First seen · 79 lines · 29 tokens per session scan A d49754fb4525
evolve-skill-library is a skill published in the GitHub repository bikeread/promethos (33 stars, last pushed 7mo ago), licensed MIT. It adds 29 tokens to every session and 701 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 2026-08-30.
evolve-skill-library compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| evolve-skill-librarybikeread/promethos | Skill | 33 | 7mo ago | MIT | A | 29 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |