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 skills add znlgis/my-opencode-deepseek-config --skill handoffgit clone --depth 1 https://github.com/znlgis/my-opencode-deepseek-configWrote 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/znlgis/my-opencode-deepseek-config/handoff)<a href="https://agentmods.dev/skills/znlgis/my-opencode-deepseek-config/handoff"><img src="https://agentmods.dev/badge/skills/znlgis/my-opencode-deepseek-config/handoff.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00059 | $0.00574 |
| Opus 5 | $0.00030 | $0.00287 |
| Sonnet 5 | $0.00012 | $0.00115 |
| Haiku 4.5 | $0.00006 | $0.00057 |
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 4d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Handoff
Compact the current conversation so a fresh agent can continue without replaying the session. Reference existing artifacts — never paste their content.
When to use
- Ending a session with unfinished work
- Task mentions "handoff", "交接", "continue later", "next session"
- Context is too large and you need to preserve current state
Structured format
Every handoff uses these exact headings, in this order:
## Goal
## Constraints & Preferences
## Progress (Done / In Progress / Blocked)
## Key Decisions
## Next Steps
## Critical Context
## Suggested Skills (optional)
Fill each with short bullets, not prose. Leave a section empty (with a single
—) rather than padding it. Omit the optional Suggested Skills heading entirely
when no skills apply.
What to include
- Goal: the end state this session was driving toward
- Progress: what's done / in progress / blocked — paths to artifacts, not content
- Key Decisions: trade-offs made, alternatives rejected, and why
- Next Steps: the concrete first actions for the next session
- Critical Context: gotchas, constraints, or facts the next agent must know
- Artifact references: paths/URLs to specs, plans, diffs, issues — never inline
What NOT to include
- Full files or large code blocks — reference paths
- Secrets (API keys, passwords, tokens, PII) — redact with
<REDACTED> - Content already captured in artifacts — the path is enough
- Session chatter irrelevant to the next agent's work
Output
Save to the OS temp directory:
| Platform | Path |
|---|---|
| Windows | $env:TEMP\opencode-handoff-YYYY-MM-DD-HHmm.md |
| Unix | $TMPDIR/opencode-handoff-YYYY-MM-DD-HHmm.md |
Agent workflow
- Collect paths of existing artifacts (specs, plans, PRs, diffs)
- Fill the structured headings with short bullets: goal, progress, decisions, next steps, critical context
- Add an optional Suggested Skills section listing skills the next session should load (omit if none)
- Write to the OS temp directory, report the path to the user
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.
- 4d ago Changed · +2 lines 2d8d78dbeec3
- 8d ago First seen · 66 lines · 59 tokens per session scan A 689eda54f7d0
handoff is a skill published in the GitHub repository znlgis/my-opencode-deepseek-config (57 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 574 once invoked, about $0.0003 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
knowledge-wiki
Manage the personal knowledge wiki. Use when the user shares articles, documents, or asks to organize knowledge; when a conversation produces insights worth preserving as structured knowledge; or when the user asks about the knowledge base.
wiki
LLM Wiki — persistent markdown knowledge base that compounds across sessions (Karpathy model).
remember
Review reusable project knowledge and decide what belongs in project memory, notepad, or durable docs.
ai-context
Project knowledge base for coding agents. Activate when: (1) starting a new session in this project, (2) encountering unfamiliar code patterns or architecture decisions, (3) user asks about project design or rationale, (4) before making significant structural changes. Contains tiered knowledge from stable design…
memory
Persistent, token-efficient project memory. When ON, maintains a .shob/memory/ folder of structured .md files so the full context of the project is NEVER lost across responses, sessions, or context compaction. Uses progressive disclosure — routes through a lightweight INDEX and loads only the files a task needs…
deja-search
Search deja before re-deriving past work: when the user refers to earlier sessions or decisions, before debugging an error, and before implementing something that may already exist. It searches this machine's own history across every AI coding tool used on it, going back further than deja itself was installed.