pause
/pause
Use to freeze current session state with rich resumption context. "Pause this", "save my place", "I need to step away". Different from native task tracking — captures full thinking state, not just todos.
- Author
- ialameh ↗
- Repository
- sift-coder ↗
- Licence
- MIT
- Last repository change
- 03.06.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 03.06.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use to freeze current session state with rich resumption context. "Pause this", "save my place", "I need to step away". Different from native task tracking — captures full thinking state, not just todos.
Triggers from the description
Original author description
Use to freeze current session state with rich resumption context. "Pause this", "save my place", "I need to step away". Different from native task tracking — captures full thinking state, not just todos.
How it works
- What gets paused
- Method
- Output shape
- Rules
- Anti-patterns
- When NOT to use
- Subagent dispatch
- Value over native task tracking
SKILL.md preview
---
name: pause
description: Use to freeze current session state with rich resumption context. "Pause this", "save my place", "I need to step away". Different from native task tracking — captures full thinking state, not just todos.
---
# pause
Session-state freeze. Saves the full context needed to resume cleanly later — including not just todos but the half-formed thoughts, the dead ends explored, the constraints discovered.
## What gets paused
1. **Open question** — what we were trying to figure out
2. **Progress** — what's done
3. **Active hypothesis** — what we currently believe
4. **Dead ends** — what we tried and why it didn't work
5. **Constraints** — discovered or stated
6. **Next concrete step** — what should happen first on resume
7. **Memory pin** — high-priority capture so resumption finds it
## Method
1. **Snapshot conversation.** Last ~20 turns, distilled.
2. **Distil the seven items above** in a structured `pause.json` at `.siftcoder/checkpoints/pause-<ts>.json`.
3. **Memory capture** with `kind: pause`, `priority: high`. Pin so it surfaces on next session.
4. **Final summary** to user — quick recap so they know what was saved.
## Output shape
Written to `.siftcoder/checkpoints/pause-<ts>.json`:
```json
{
"id": "pause-<ts>",
"createdAt": "<ISO>",
"openQuestion": "...",
"progress": ["...", "..."],
"hypothesis": "...",
"deadEnds": [
{ "tried": "...", "result": "..." }
],
"constraints": ["..."],
"nextStep": "...",
"memoryId": "summary:..."
}
```
User-facing:
```
Paused at: <ts>
Open question: <one line>
Progress so far:
✓ <item>
✓ <item>
Currently believing: <hypothesis>
Resume with: /siftcoder:continue <id>
or /siftcoder:continue (most recent)
```
## Rules
- **Capture dead ends explicitly.** Otherwise resumption re-explores them.
- **Constraint capture is mandatory.** "Cannot use X because Y" is what's hardest to recover from memory alone.
- **Memory pin priority high.** Routine summaries are low-priority; pauses are high.
- **Single concrete next step.** Not a list — one specific action to start on resume.
## Anti-patterns
- Pause that just says "continue from where we left off" (no info)
- Listing 10 next-steps (paralyses on resume)
- Forgetting dead ends (high-cost rediscovery)
- Long prose summary nobody reads on resume
## When NOT to use
- Trivial work (just remember it)
- End of project (use `/handoff` instead — different shape)
- Session that's a scratch exploration — losing it is fine
## Subagent dispatch
- None — pause is a synthesis of current conversation
- Memory MCP for the pin
## Value over native task tracking
Native task tracking captures todos. Pause captures **why** the todos exist — hypothesis, dead ends, constraints. The thinking-state IS the value.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · ialameh/sift-coder | 100 | MIT |
| wrg32786/aigent-os | 100 | MIT |
| shakacode/agent-workflows | 100 | MIT |
| character-ai/larch | 95 | MIT |
| Everyone-Needs-A-Copilot/claude-copilot | 90 | MIT |
Database snapshot: 13.11.2026 · package version b4696e6e94ed (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100context-saver
Write the current session's in-flight work state to a human-readable handoff doc on disk (docs/context/CONTEXT-{topic}.md) so a future chat session — or another person — can pick up exactly where this one stops. Distinct from Claude Code's built-in memory system, which captures… - 100context-capsule
Write a resume-ready capsule of the current session (objective, what it is waiting on, the next concrete action) so a /clear or an auto-compact can pick up exactly where this left off. Write it before compaction; nothing writes one for you at that point. Use for a checkpoint… - 100dbs-save
把当前诊断的关键状态存到本地,下次回来可以接着用。 触发方式:/dbs-save、/存档、「保存这次诊断」「记下来」「这个结论留着」 Save the current diagnosis state to disk for cross-session recall. Trigger: /dbs-save, "save this diagnosis", "remember this - 100ctx-pause
Pause context hooks for this session. Use when context nudges aren't needed for the current task. - 100ph
Prepare session handoff for continuation in a new conversation. Use when 'prepare handoff', 'save progress', 'session handoff', 'I need to stop', 'prepare for next session', 'hand off', 'write handoff', or when the user wants to pause mid-implementation and resume later in a fresh context window. - 100ph
Prepare session handoff for continuation in a new conversation. Use when 'prepare handoff', 'save progress', 'session handoff', 'I need to stop', 'prepare for next session', 'hand off', 'write handoff', or when the user wants to pause mid-implementation and resume later in a fresh context window.