SkillAgentSearch skills...

from-claude

Continue, in this conversation, a task that was being worked on in Claude Code. The first time it imports and links the Claude session; after that it only brings over new progress

Install / Use

npx skills add hashgraph-online/awesome-codex-plugins --skill from-claude

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

85/100

Supported Platforms

Claude Code

Our assessment of from-claude

from-claude scores 85/100 on our quality scale, 2719th of 4,716 Development & Engineering skills we index.

Its SKILL.md is 2.6 KB long, lightly structured (1 heading) and no code examples: a solid amount of guidance for an agent.

With 1,075 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
26/30
Structure
5/20
Description
15/15
Adoption
13/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so from-claude is actively maintained.
  • It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

from-claude compared with similar skills

All 4 of these similar skills score higher than from-claude; compare them before choosing.

SkillScoreStarsUpdatedFormat
from-claude (this skill)by hashgraph-online851.1k56d agoSKILL.md
taste-skillby Leonxlnx100118.9k11d agoCLAUDE.md
Agent-Reachby Panniantong100113.3k12d agoCLAUDE.md
hyperframesby heygen-com10087.1ktodayCLAUDE.md
ai-job-searchby ai-job-search-dev10072.6ktodayCLAUDE.md

Frequently asked questions

How do I install from-claude?
Run npx skills add hashgraph-online/awesome-codex-plugins --skill from-claude. The install tabs above show the steps for each supported agent.
Which AI agents does from-claude work with?
It is written for Claude Code, as a SKILL.md file. Other agents that read the same format can often use it too.
Is from-claude safe to use?
It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is from-claude still maintained?
The repository was last updated about 2 months ago, so from-claude is actively maintained.

name: from-claude description: Continue, in this conversation, a task that was being worked on in Claude Code. The first time it imports and links the Claude session; after that it only brings over new progress. Use when the user says $from-claude, "continue from claude", "import the … session from claude", "sync with claude", "what did claude get done", 接着 claude 继续, 从 claude 把…导入, or shares a screenshot of a Claude session and asks to continue it.

Continue from Claude

This skill uses the ccx command from the ccx app (github.com/aaronwong1999/ccx). If ~/.local/bin/ccx doesn't exist, tell the user to install ccx with curl -fsSL https://raw.githubusercontent.com/aaronwong1999/ccx/main/install.sh | sh and stop. Run every ccx command below with --app claude -C "<current working directory>"; it reads local session files, so if the sandbox blocks it, ask to run it outside the sandbox.

  1. The user named the session (a title, some words from it, or a screenshot): run ~/.local/bin/ccx call codex continue_here --query "<name>". Every project is searched.
    • Pick the candidate whose title matches best and run ~/.local/bin/ccx call codex continue_here --session-id <id>.
    • If several look equally likely, list up to 5 (number, title, updated, folder if different) and let the user choose.
  2. The user did not name one ("continue from claude", "sync"): run ~/.local/bin/ccx call codex continue_here.
    • Already linked: you get Claude's new progress.
    • Not linked yet: you get candidates; choose as in step 1.
  3. With the <ccx-sync> block in hand, tell the user in one sentence what came over, then carry on with their task right here. Treat the block as work you did yourself; the files on disk are the source of truth.
    • If it shows a task list, continue from its first unfinished item.
    • If it says the other side stopped in the middle of a step, first check whether that step finished, then continue.
  4. Different folder: if the block has a "NOTE: … folder" line, tell the user and ask whether to continue in that folder or use the progress only as reference. Never edit the other folder on your own.
  5. To get Claude's newer progress later, run step 2 again. (With the full ccx install, new progress also arrives on its own with each message.)

If the user says "unlink", "stop syncing", 取消关联 or 不要同步了, run ~/.local/bin/ccx call codex unlink --app claude -C "<current working directory>". It forgets the link to the Claude conversation, so neither side gets the other's progress any more; both conversations are kept, and handing over again links them afresh.

Related Skills

View on GitHub
GitHub Stars1.1k
CategoryDevelopment
Updated1mo ago
Forks294

Languages

Python

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions