菜单
Resume work from the long-session artifacts (claude-progress.txt + spec.md + features.json). Reads the append-only log, the living spec, and the testable-requirements list, then briefs the current session.
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 | Resume work from the long-session artifacts (claude-progress.txt + spec.md + features.json). Reads the append-only log, the living spec, and the testable-requirements list, then briefs the current session. |
| when_to_use | Reach for this at the start of a new session to pick up where the previous one left off, or after a `/compact` when context state has been lost. Do NOT use for writing resume artifacts — use `/progress-log` or `/feature-list` instead; session-resume is the read-side briefer. |
| disable-model-invocation | true |
| allowed-tools | ["Read","Bash","Glob","Grep"] |
| logical | briefing surfaces last 3 progress entries, spec tail, and features.json status |
Reads the long-session artifacts; outputs a briefing so work continues without re-exploration.
Check init.sh. If init.sh exists at repo root and is executable, remind the user:
init.sh is present. If the dev env isn't up, run: bash init.sh
Read claude-progress.txt tail. Entries are blank-line-separated blocks (shape defined by /progress-log). Show the last 3 entries verbatim.
Read .claude/spec.md. If present, show the tail (last 20 lines or the last delta block, whichever is smaller).
Read .claude/features.json. Summarize: N pending, M in_progress, K done. List any in_progress items by id + description.
Git state.
git status --short
git log --oneline -5
Detect test command.
package.json scripts.testcomposer.json scripts.testMakefile test targetpytest / cargo test / go test
Note it; do NOT auto-run.Emit briefing in this exact shape:
## Session Briefing
**Last progress entry:** <date> — <topic>
<last entry tail>
**Features:** N pending · M in_progress · K done
In progress: F3 <desc>, F5 <desc>
**Spec tail:**
<last spec delta>
**Git:** <N commits today>, <M uncommitted files>
**Tests:** <command detected, or "no command detected">
Ready. What's next?
init.sh at repo root; remind the user to run bash init.sh if presentclaude-progress.txt verbatim.claude/spec.md tail (last 20 lines or last delta block).claude/features.json (pending / in_progress / done counts; list in-progress items)git status --short, git log --oneline -5Ready. What's next?/progress-log, /living-spec, /feature-list, /init-sh.surface-progress.sh (SessionStart) — that hook previews; this skill produces the full briefing on demand.No long-session artifacts found. Start with Plan mode + /feature-list + /living-spec..claude/features.json malformed → report parse error; skip that section but continue.Use when the user describes a multi-step feature or refactor and you need to decide whether to handle it solo, dispatch a parallel `/fan-out`, or run a `/worktree-team` planner→generator→reviewer pipeline. Outputs a routing recommendation with the reasoning behind it.
Use whenever the user asks to apply the same operation to many independent files or components — bulk migration, batch refactor across controllers, parallel exploration of unrelated subsystems. Dispatches one subagent per file with a shared prompt template, then collects the results.
Use to toggle the safe-mode flag at .claude/safe-mode. The flag is written automatically by consecutive-failure-guard once tool failures hit FORGE_SAFE_MODE_THRESHOLD (default 5), and block-destructive.sh denies every Bash/Write/Edit while it exists. /safe-mode off clears the flag, logs the exit, and prompts /postmortem; /safe-mode on enters manually; /safe-mode status reports current state.
Run the 5-stage Token Transformation Pipeline (TRAE §5.2.2) over the current session state — Collection, Ranking, Compression, Budgeting, Assembly. Emits a concrete next-action recommendation (/compact, /lean-md, or /progress-log + fresh session) instead of a generic warning.
Audit or scaffold a CLAUDE.md against the Karpathy 4-section structure (Think Before Coding · Simplicity First · Surgical Changes · Goal-Driven Execution). Companion to /lean-md — structure audit, not size trim.
Print every policy enforcement point declared in plugins/diagnostics/registry/policies.json — id, verdict, plugin, hook event, severity, bypass — grouped by verdict. Single discoverable index of what the harness blocks, anchors, nudges, or logs at runtime.