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.
npx agentmods add skills/mehrad-dm/mastermind/handoffnpx skills add mehrad-dm/mastermind --skill handoffgit clone --depth 1 https://github.com/mehrad-dm/mastermindWrote 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/skills/mehrad-dm/mastermind/handoff)<a href="https://agentmods.dev/skills/mehrad-dm/mastermind/handoff"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/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 | $0.00079 | $0.00623 |
| Opus 5 | $0.00039 | $0.00311 |
| Sonnet 5 | $0.00016 | $0.00125 |
| Haiku 4.5 | $0.00008 | $0.00062 |
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 6d 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MasterMind: Handoff
Context is the fundamental constraint (~/.mastermind/engineering/core/agent-loop.md); a long task shouldn't lose its thread to a
reset. Capture just enough to resume cold: high signal, low tokens.
Capture
- Goal: the outcome being built, in one line.
- State: Done / In-progress / Next, as a short checklist. Be honest about what's not verified.
- Key decisions & why: the non-obvious choices and their rationale (so they aren't relitigated).
- Map: where the relevant code/files/config live; the entry points.
- Gotchas: traps, failing things, env quirks, anything that bit you.
- How to resume & verify: the exact command(s)/check to pick up and confirm green.
Rules
Summarize and link, point to detail rather than pasting it. Write it where the next session
will look (.mastermind/HANDOFF.md, a scratch file, or the issue). Prune it when the work completes.
Output
A tight .mastermind/HANDOFF.md: goal, state checklist, decisions, file map, gotchas, resume+verify steps.
The ledger: for work that outlives one context window
A handoff is written once, at the pause. A ledger runs the whole way through: an append-only record of what has actually been completed, so a long task can lose its thread without losing its place. Start one when the work will be summarized more than once.
- First line names the work: the plan, spec, or issue it belongs to. A ledger nobody can match to a plan is a page of orphaned sentences.
- Append one line as each step completes: what was done, and the check that proved it. Earlier entries are never rewritten or tidied: it is a record of what happened, not a plan of what will.
- Conversation memory does not survive compaction: trust the ledger and the git log over your own recollection, most of all when you feel certain. Re-doing finished steps is the costliest failure in long-horizon work, and it always arrives wearing "I remember doing that." Read the record, then act.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 43 lines · 0 tokens per session scan A d0f05389b0ce
handoff is a skill published in the GitHub repository mehrad-dm/mastermind (24 stars, last pushed 6d ago), licensed MIT. It adds 79 tokens to every session and 623 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
agentbook-version-control
Coordinate Git and Cross-Session Memory as version control for agent work. Use when Codex needs AgentBook status, checkpoints, history, resume or handoff, provenance, isolated agent approaches, semantic comparison, or an evidence-backed game-development tuning and playtest iteration with an explicit keep or revert…
csm-continuity
Use the complete native Cross-Session Memory (CSM) runtime for continuity, memory governance, living state, beliefs, self-model, AgentBook, checkpoints, context cache, goals, work ledger, compaction, re-entry, or handoff across Codex tasks.
csm-governance
Safely deduplicate, review, merge, and archive Cross-Session Memory. Use when memory needs cleanup, dedup, merge, archive-candidate review, or recall-quality assessment.
wisdom
Primary runtime memory skill for institutional knowledge, operational facts, patterns, and decisions. Use for ALL knowledge queries including deployment, config, and infrastructure.
vibebox
Use this skill when an AI coding task should consult VibeBox memory before work as an auto-intervention memory layer for meaningful repository work; retrieve active guidance when available, apply it as context, and capture meaningful outcomes with userRequest plus AI-agent structured memory candidates unless the user…
remember
Triage a session's findings across memory surfaces — classify each item and route it to the right destination (project memory, CLAUDE.md/AGENTS.md, or session notes) rather than dumping everything into one store.