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 lycfyi/yskills --skill human-context-rebuildgit clone --depth 1 https://github.com/lycfyi/yskillsWrote 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/lycfyi/yskills/human-context-rebuild)<a href="https://agentmods.dev/skills/lycfyi/yskills/human-context-rebuild"><img src="https://agentmods.dev/badge/skills/lycfyi/yskills/human-context-rebuild/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/lycfyi/yskills/human-context-rebuild"><img src="https://agentmods.dev/badge/skills/lycfyi/yskills/human-context-rebuild.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00214 | $0.00870 |
| Opus 5 | $0.00107 | $0.00435 |
| Sonnet 5 | $0.00043 | $0.00174 |
| Haiku 4.5 | $0.00021 | $0.00087 |
Grade A, and why
human-context-rebuild 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 9d 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Human Context Rebuild
Purpose
Drop the user back into the session in under 30 seconds of reading. They forgot. Their job is to scan and resume — not to re-read a transcript.
Produce one short recap. No preamble, no "great question", no offering to do more.
Output structure
Use exactly these sections, in this order. Skip a section only if there's genuinely nothing to put in it (don't pad).
**现在在做 / Now**: <1 line — the active task in concrete terms>
**为什么 / Why**: <1 line — the trigger or motivation the user gave>
**目标 / Goal**: <1 line — the success condition>
**进展 / Done**: <bullet list, ≤4 items, most recent first>
**下一步 / Next**: <1–2 lines — the very next action or open decision>
**Open questions** (only if blocking): <bullets>
Total length: aim for under 150 words. Hard cap 250.
Language
Match the session's dominant language. If the user has been mixing Chinese and English, mix in the recap too — keep technical terms (file paths, function names, statuses) in their original form. Don't translate identifiers.
What to include
- Concrete anchors: file paths, function names, branch names, decisions the user explicitly confirmed. These are what the user's brain latches onto faster than prose.
- Decisions made, especially scope changes ("user dropped quota, kept only AI judge").
- The most recent meaningful action: what the last tool call or agent run actually produced.
What to leave out
- Tool-call mechanics (which agent ran, how many files were grep'd).
- Detailed code or SQL — link to the file path instead.
- Praise, hedges, "let me know if…", or offers to continue. The user can see the next step; they'll drive.
- Things the user already said in this turn. Don't echo their own message back.
- Re-justification of past decisions. State them as settled.
Edge cases
- Very new session (under ~3 turns): say so directly. Example:
我们才刚开始 — 你给的方向是 X, 还没动手. 要不要先 ___? - Session changed topics mid-way: recap the current topic only. Mention the prior topic in one line if it's still open:
(早些时候在弄 Y, 暂停了.) - Just finished a big chunk of work: lead Progress with what shipped, then Next with verification or the follow-up commit.
- In the middle of a blocked tool call or pending question: put the blocker in Open questions and make Next be "answer the question above".
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.
- 9d ago First seen · 60 lines · 214 tokens per session scan A 6983a00c877d
human-context-rebuild is a skill published in the GitHub repository lycfyi/yskills (23 stars, last pushed 14d ago), licensed MIT. It adds 214 tokens to every session and 870 once invoked, about $0.0011 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
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…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…