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 jon-devlapaz/tink-skills --skill seed-megit clone --depth 1 https://github.com/jon-devlapaz/tink-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/jon-devlapaz/tink-skills/seed-me)<a href="https://agentmods.dev/skills/jon-devlapaz/tink-skills/seed-me"><img src="https://agentmods.dev/badge/skills/jon-devlapaz/tink-skills/seed-me/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/jon-devlapaz/tink-skills/seed-me"><img src="https://agentmods.dev/badge/skills/jon-devlapaz/tink-skills/seed-me.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 | $0.00000 | $0.06020 |
| Opus 5.5 | $0.00000 | $0.02408 |
| Sonnet 5.5 | $0.00000 | $0.01204 |
| Haiku 4.5 | $0.00000 | $0.00602 |
Grade A, and why
seed-me 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 today.
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 — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Seed Me
Resolve inspectable facts yourself; reserve questions for consequential user judgment. Keep actual answers separate from recommendations. Every claim carries its receipt; every question names its owner.
1. Triage the ask and extract context
- Open an interview for any input the user wants formalized — a plan, a loose brainstorm, braindump, hunch, or idea. Every input takes the same path: shape a working draft first, then grill it. Ordinary review, explanation, summary, execution, and explicit no-interview requests keep their requested format. Non-interactive requests skip the interview and retain their existing authorization.
- Do not read the ledger reference yet: the size gate below decides whether the interview needs a ledger at all. On the Full path, start the session when the user confirms the goal (see Session lifecycle below); until then the working draft lives in chat. Maintain the decision ledger throughout; visible questions are a projection of this ledger.
- Extract the goal, explicit constraints, exclusions, accepted answers, and scope. Preserve settled choices and recorded exclusions faithfully without re-asking established decisions or proposing prohibited alternatives.
Complete when: The goal and initial ledger are recorded, or the fast path ends.
Shape the working draft
Distill whatever arrived into a working draft before any grill turn: candidate goal, candidate proposed outcome, and candidate options with tradeoffs. Keep the graph empty until goal confirmation; retain choices already explicit in the input and record them when the confirmed nodes are published. Anything uncertain, assumed, or missing is labeled PROVISIONAL assumption, never an answer or a claim. The draft lives in chat until the goal is confirmed: it creates no session and no ledger nodes, and needs no publish. Provisional lines need no 📜 and spend no ungrounded budget; they are scaffolding for the user to correct, and a braindump simply yields more of them than a polished plan does. Present the draft in one tight block and ask what to keep, cut, or reshape. Confirmed lines become ledger nodes; rejected lines are dropped, not parked. A general confirmation accepts the investigation goal, not its candidate solutions; record only explicit choices as accepted answers. Only then does the frontier loop start.
What ships with it
15 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.
- assets/ledger-view.html 36 KB
- assets/ledger.json 426 B
- references/agent-mode.md 6.9 KB
- references/engram-council.md 3.9 KB
- references/lean-path.md 4.1 KB
- references/ledger-transitions.md 19 KB
- references/nasa-spike/conops.md 960 B
- references/nasa-spike/decision-analysis.md 1008 B
- references/nasa-spike/PROMPT.md 3.1 KB
- references/nasa-spike/requirements-taxonomy.md 1.5 KB
- references/nasa-spike/run_eval_contracts.py 1.9 KB runs code
- scripts/engram_brief.py 4.4 KB runs code
- scripts/engram_council.py 6.1 KB runs code
- scripts/session.py 20 KB runs code
- scripts/viewer.py 3.0 KB runs code
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.
- today Changed · +16 lines 4b7889907cdc
- yesterday Changed · +84 lines 648e473dd97b
- 3d ago Changed · +61 lines 9748e284632e
- 4d ago First seen · 232 lines · 0 tokens per session scan A 68374a3cc513
seed-me is a skill published in the GitHub repository jon-devlapaz/tink-skills (16 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 6,020 tokens. 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 2026-09-27.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…