save-handoff
The save-handoff command creates a structured markdown file in `.aspens/sessions/` containing task summary, current progress, affected files, risks, and next steps, enabling future Claude sessions to resume work efficiently. Use this when pausing extended projects to preserve context and maintain continuity across separate coding sessions.
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/aspenkit/aspens/HEAD/.claude/commands/save-handoff.md -o ~/.claude/commands/save-handoff.mdsave-handoff.md
Write a structured handoff file to `.aspens/sessions/` so a future Claude session can continue this work.
## Steps
1. Generate a timestamp: `YYYY-MM-DDTHH-MM-SS` (replace `:` and `.` with `-`).
2. Write a markdown file to `.aspens/sessions/<timestamp>-claude-handoff.md` with this structure:
```md
# Claude save-tokens handoff
- Saved: <ISO timestamp>
- Reason: user-requested
- Working directory: <cwd>
## Task summary
<1-3 sentences: what you were working on and why>
## Current state
<What's done, what's in progress, what's blocked>
## Files touched
<List of files created, modified, or deleted in this session>
## Risks and open questions
<Anything the next session should watch out for>
## Next steps
<Concrete next actions to continue this work>
```
3. Update `.aspens/sessions/index.json` with `{ "latest": "<relative path>", "savedAt": "<ISO>", "reason": "user-requested" }`.
4. Confirm the handoff was saved and print the file path.LLM-powered injection of project context into installed agent templates via `aspens customize agents`
>
Core conventions, tech stack, and project structure for aspens
Claude/Codex CLI execution layer — prompt loading, stream-json parsing, file output extraction, path sanitization, skill file writing, and skill rule generation
Top-level Commander wiring, welcome screen, missing-hook warning, CliError exit handling, and the public programmatic API surface
Multi-target output system — target abstraction, backend routing, content transforms for Codex CLI and future targets
Context health analysis — freshness, domain coverage, hub surfacing, drift detection, LLM-powered interpretation, and auto-repair for generated agent context
Incremental skill updater that maps git diffs to affected skills and optionally auto-syncs via a post-commit hook