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/alivecontext/alive/alive-capturenpx skills add alivecontext/alive --skill alive-capturegit clone --depth 1 https://github.com/alivecontext/aliveWrote 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/alivecontext/alive/alive-capture)<a href="https://agentmods.dev/skills/alivecontext/alive/alive-capture"><img src="https://agentmods.dev/badge/skills/alivecontext/alive/alive-capture.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 | $0.00021 | $0.00489 |
| Opus 5 | $0.00010 | $0.00244 |
| Sonnet 5 | $0.00004 | $0.00098 |
| Haiku 4.5 | $0.00002 | $0.00049 |
Grade A, and why
alive-capture 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.
What it actually says
Capture Context
External content arrives -- email, transcript, file, research output. Route it into the system.
Detect Content Type
- Pasted text, forwarded message -> extract and store
- File path mentioned -> read and classify
- Research results from a cron or web search -> structure and route
- Screenshot -> store in bundle raw/
Route Decision
This looks like [type]. Route it?
1. Add to [active bundle] raw/
2. Create new bundle for it
3. Route to [suggested walnut]
4. Just stash it for now
Capture Process
- Name the file. Date-prefixed, descriptive:
2026-04-07-witcheer-feedback.md - Store in bundle raw/. Write the content to the active bundle's
raw/directory. - Update manifest. Add to
sources:incontext.manifest.yaml:
sources:
- path: raw/2026-04-07-witcheer-feedback.md
description: Telegram feedback on ALIVE x Hermes spec
type: document
date: 2026-04-07
- Extract stash items. Any decisions, tasks, insights, people mentions, or sharp quotes get stashed.
- Rename garbage filenames.
CleanShot 2026-02-23...becomes2026-02-23-descriptive-name.ext. Preserve original in manifest asoriginal_filename:.
If No Bundle Active
Ask which walnut this belongs to, then offer to create a bundle or add to an existing one.
If No Walnut Active
Check 03_Inbox/ for unrouted files. Enter inbox scan mode -- present each file with routing suggestions.
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 First seen · 65 lines · 21 tokens per session scan A 01fd5c99cbbd
alive-capture is a skill published in the GitHub repository alivecontext/alive (127 stars, last pushed 9d ago), licensed MIT. It adds 21 tokens to every session and 489 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 2026-08-30.
Other skills, from other repositories
cco-overhead
Audit the fixed context overhead every session starts with — system prompt, MCP tools, agents, CLAUDE.md, memory — measured from real transcript usage.
handoff
Create a session briefing for the next session — auto-loaded, then archived. Use when the user is wrapping up a session, says "save this for next time", "hand off", "brief the next session", or wants to ensure continuity before stopping work.
forget
Remove entries from remembered items. Use when the user wants to delete, remove, or clear a previously remembered item, or says "forget this", "remove that memory", or "clear my preferences".
recall
Inject filtered context from previous sessions. Use when the user says "recall", "what happened last session", "what did we work on", "inject previous context", or when you need context from a prior session to continue work. Also useful mid-session when the user references prior work you don't have context for.
remember
Save something to memory for future sessions. Use when the user says "remember this", "save for later", "don't forget", or wants to persist any preference, fact, lesson, or note across sessions.
status
Check plugin health and diagnose issues. Use when the user asks about mneme health, memory status, plugin errors, wants to troubleshoot memory issues, or sees error warnings at session start.