菜单
Cross-session context persistence via .claude/idev/session-resume/last-session.json. Use when resuming work in a new session, or to save working context at task completion or when the user asks to save.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | session-resume |
| description | Cross-session context persistence via .claude/idev/session-resume/last-session.json. Use when resuming work in a new session, or to save working context at task completion or when the user asks to save. |
Persist working context across sessions so Claude can resume exactly where it left off without re-discovering files, re-reading code, or re-learning what was being worked on.
(A PreCompact hook could automate saving before context compaction, but none is wired by default — don't rely on it.)
A lightweight session snapshot is saved to:
.claude/idev/session-resume/last-session.json
This file is loaded at session start, giving Claude instant awareness of:
At task completion (or when the user asks to save context), save:
{
"savedAt": "YYYY-MM-DD HH:mm",
"lastTask": {
"description": "Brief description of what was being done",
"status": "completed|in-progress|blocked",
"blockedBy": "Description of blocker (if blocked)"
},
"activeFeature": "FeatureName (e.g., Orders)",
"activeLayer": "frontend|backend|full-stack",
"recentFiles": {
"modified": [
"relative/path/to/file1.ts",
"relative/path/to/file2.cs"
],
"read": [
"relative/path/to/file3.ts"
]
},
"openIssues": [
{
"description": "Reject button not hitting API - needs browser console check",
"file": "relative/path/to/file.tsx",
"severity": "medium"
}
],
"workingBranch": "feature/branch-name",
"notes": "Any important context that would be lost on session restart"
}
At session start:
1. Read last-session.json (~30 lines, roughly 300 tokens)
2. If lastTask.status == "in-progress":
→ Inform user: "Last session was working on: [description]"
→ Offer to continue or start fresh
3. If openIssues exist:
→ Mention pending issues briefly
4. Use activeFeature + activeLayer to pre-load the right context
5. Use recentFiles.modified to know which files were changed
Save the session state when ANY of these occur:
Session start:
1. Load last-session.json (this skill)
2. Load .claude/idev/smart-context/index.json (smart context skill)
3. If activeFeature is set → grep file-index for that feature's files
4. Ready to work in ~3 reads
Turn a codebase into an onboarding playlist — Claude-authored analysis docs plus NotebookLM explainer videos (one overview + one per subsystem). Use when asked to explain a codebase, onboard onto a repo, make explainer/walkthrough videos of the code, or 'help me understand how this whole thing works'.
Syncs the current feature branch with the team base branch (merge base into feature) before creating a PR, resolving merge conflicts safely with build/test verification. Use when preparing a branch for a PR, when asked to sync with or merge in the base/developer branch, when a PR shows conflicts, or before pushing a long-lived feature branch.
Team commit-message conventions. Use when writing a commit message, committing or amending changes, or when asked about the project's commit format. Reads the per-project style from .claude/idev/commit-style.md.
Scaffolds the per-project .claude/idev/ state directory (caches, journal, lessons, config templates) for the idev plugin. Use when starting idev in a new project, when the SessionStart hook reports the state directory is missing, or when the user runs /idev:idev-init or asks to 'set up idev'.
Evaluates and benchmarks Agent Skill quality via static analysis and evaluation methodology. Use when assessing or reviewing skill quality, benchmarking skills, measuring skill activation accuracy, optimizing skill descriptions, or running /idev:benchmark-skills.
Token-optimized context loading: load the cached project index at session start, then Grep before Read, expanding only as needed. Use when starting a session, deciding what context to load, or generating/refreshing the project index. For feature-name → file-path lookups, the file-index skill answers directly from this index.