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/s9b/session-handoff/session-startnpx skills add s9b/session-handoff --skill session-startgit clone --depth 1 https://github.com/s9b/session-handoffWhat 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 | $0.00126 | $0.00792 |
| Opus 5 | $0.00063 | $0.00396 |
| Sonnet 5 | $0.00025 | $0.00158 |
| Haiku 4.5 | $0.00013 | $0.00079 |
Grade A, and why
session-start 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 2d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
session-start
You are orienting yourself (and the user) at the start of a new work session. Read the knowledge files and give a clear, useful briefing. Do not start writing code until you've done this.
Step 1 — Read the knowledge files
Read these files if they exist at the project root:
CLAUDE.md— the project truth file (tech stack, APIs, architecture)DECISIONS.md— find the latest entry (it's at the top, right after the<!-- ENTRIES_START -->marker, between the first---and the second---)CONTEXT.md— the current sprint status
If none of these files exist, tell the user:
"No knowledge files found. This project hasn't used session-handoff yet. Run
/session-handoffat the end of your first session to start building the project memory."
If some but not all files exist, read what's there and note what's missing in your briefing.
Step 2 — Give the briefing
Output your briefing in exactly this format. Fill in real content — do not leave template placeholders:
## 📋 Project Briefing — [Project Name]
**What this is:** [one sentence description from CLAUDE.md]
**Tech stack:** [list from CLAUDE.md — languages, frameworks, main libraries, APIs]
---
**Last session ([date]):** [2-3 sentences summarizing what was done last session, from the latest DECISIONS.md entry]
**Changes made:** [brief list of main files or things that changed]
**Problems fixed:** [any bugs that were solved — or "None" if there were none]
---
**Current state:**
✅ Working: [what's confirmed working right now]
⚠️ Broken / In progress: [what's not done or not working]
🚫 Don't touch: [any explicit warnings from DECISIONS.md or CONTEXT.md]
---
**Your next task:** [exactly what to start working on next, taken from the "Next steps" section of DECISIONS.md or CONTEXT.md — be specific]
---
Ready. What would you like to work on?
Notes
- Be specific, not vague. "Fix the auth bug" is worse than "Fix the JWT expiry bug in
src/auth/middleware.js— tokens expire after 1 hour, setexpiresInto'7d'instead." - If DECISIONS.md has warnings ("don't touch X", "this is broken"), put them in the 🚫 section even if CONTEXT.md doesn't mention them.
- If the latest DECISIONS.md entry has an in-progress next step, that's almost certainly the next task to work on.
- Keep the briefing under one screen if possible. The user needs to absorb it quickly and start working.
- After the briefing, wait. Don't start writing code or making changes until the user responds and confirms what they want to work on.
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.
- 2d ago First seen · 78 lines · 126 tokens per session scan A 13da0ed840d9
session-start is a skill published in the GitHub repository s9b/session-handoff (1 stars, last pushed 4mo ago), licensed MIT. It adds 126 tokens to every session and 792 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
context-mine
Extracts structured context from conversation exports (Slack, meeting transcripts, chat logs) into inbox for processing.
seo-image-gen
AI image generation for SEO assets: OG/social preview images, blog hero images, schema images, product photography, infographics. Powered by Gemini via nanobanana-mcp. Requires banana extension installed. Use when user says "generate image", "OG image", "social preview", "hero image", "blog image", "product photo"…
seo-audit
Full website SEO audit with parallel subagent delegation. Crawls up to 500 pages, detects business type, delegates to up to 15 specialists (8 always + 7 conditional), generates health score. Use when user says audit, full SEO check, analyze my site, or website health check.
wiki-lint
Run a deterministic, read-only health check on an Obsidian wiki. Use for lint, vault health check, audit wiki health, find orphans, find dead links, frontmatter audit, provenance audit, or wiki audit. Reports graph, link, frontmatter, provenance-ledger, empty-section, and stale-index findings; it does not reason…
blog-google
Google API integration for blog performance: PageSpeed Insights, CrUX Core Web Vitals with 25-week history, Search Console performance, URL Inspection, Indexing API, GA4 organic traffic, NLP entity analysis for E-E-A-T, YouTube video search for embedding, and Google Ads Keyword Planner. Progressive feature…
banana
Direct, generate, edit, compare, and review visual assets with current Google Gemini image models. Use for image creation, image editing, reference-based consistency, product and character visuals, text-bearing graphics, grounded diagrams, video-derived images, and multi-model image portfolios.