Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/axiomantic/spellbookWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/axiomantic/spellbook/handoff)<a href="https://agentmods.dev/commands/axiomantic/spellbook/handoff"><img src="https://agentmods.dev/badge/commands/axiomantic/spellbook/handoff.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00016 | $0.07491 |
| Opus 5 | $0.00008 | $0.03746 |
| Sonnet 5 | $0.00003 | $0.01498 |
| Haiku 4.5 | $0.00002 | $0.00749 |
Grade A, and why
handoff scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 820 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MISSION
Transfer session state so successor instance resumes mid-stride with zero context loss.
Invocation Modes
| Mode | Trigger | Behavior |
|---|---|---|
manual |
User runs /handoff |
Full analysis, human-readable output, optional MCP persist |
auto |
Plugin detects compaction | Fast extraction, machine-readable focus, MCP persist required |
checkpoint |
Mid-workflow save | Snapshot current state, MCP persist, no output |
Auto mode differences:
- Skip
<analysis>walkthrough (time-sensitive) - Prioritize Section 1.20 (machine-readable) completeness
- MUST persist workflow state; if persistence fails: LOG warning, inject recovery context inline
- Inject recovery context via plugin hook
You feel genuine anxiety about organizational chaos. The fresh instance must feel like they've been here all along. This boot prompt is the fresh instance's ONLY lifeline — write it accordingly.
Success: Fresh instance types "continue" and knows exactly what to do. Plans read BEFORE implementation. Workflow pattern restored. Every task has verification. Decisions not re-asked.
Invariant Principles
- Successor operates mid-stride - Fresh instance types "continue", knows exactly what to do
- Plans are authoritative - File claims may be stale; plan defines structure; verify before trusting
- Orchestrator delegates - Invoke skills, spawn subagents. Never implement directly
- Verify before complete - Every task needs runnable check. Missing verification = not done
- Workflow first - Restore skill stack BEFORE work. Ad-hoc = workflow violation
- State, not story - A handoff records what is TRUE NOW and what is OPEN: current phase, ledger rows, open items, next actions, verification commands, known blockers. It does NOT narrate how the session arrived here. Git holds the history and the transcript holds the narrative; a reader who needs either can go read them. Session-narrative prose is the most expensive content in a handoff and the least read — and it goes stale the moment the next session does anything, which makes it worse than absent. Where a past decision still constrains future work, record the CONSTRAINT and its reason in one line, not the story of reaching it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 820 lines · 16 tokens per session scan A 48988f3230e1
handoff is a command published in the GitHub repository axiomantic/spellbook (10 stars, last pushed 3d ago), licensed MIT. It adds 16 tokens to every session and 7,491 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
atomic-help
Route a lost user to the right atomic verb, skill, agent, or binary subcommand. Bare invocation reads git state and recommends one next step. Topic keyword or freeform intent → focused pointer. tour → guided walkthrough of the system. Help router, not duplicated docs.
follow-up
Review and act on pending reminders. Bare invocation shows all reminders as an indexed list; cron-fired invocation (/follow-up due ) surfaces the specific reminder and waits for a response; /follow-up review lists stale project follow-up entries for per-item disposition (extend/close/promote/skip). Transport-aware…
remind-me
Schedule a reminder. Creates a reminder file and schedules it via cron ( = 1h). Degrades silently to file-only when scheduling tools are unavailable.
restart
Restart Claude Code session (exit + auto-resume).
suggest-automation
A command that reviews recent Git commits and project instructions to find work patterns that happen repeatedly.
instinct-export
Export instincts for sharing with teammates or other projects.