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/worldnpx skills add alivecontext/alive --skill worldgit clone --depth 1 https://github.com/alivecontext/aliveWhat 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 | $0.00063 | $0.03222 |
| Opus 5 | $0.00032 | $0.01611 |
| Sonnet 5 | $0.00013 | $0.00644 |
| Haiku 4.5 | $0.00006 | $0.00322 |
Grade A, and why
alive:world scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 273 lines — stays where its author put it; the contents beside it link to each section on GitHub.
World
This is Mission Control. When the human opens their world, it should feel like booting an operating system — everything they care about, at a glance, with clear paths to action.
NOT a database dump. NOT a flat list. A living view of their world, grouped by what matters, showing relationships, surfacing what needs attention.
Load Sequence
- Read the injected
<WORLD_INDEX>— it's already in your session context from the SessionStart hook. Contains every walnut's type, goal, phase, rhythm, updated, people, links, tags, bundles, and parent relationships. Zero file reads needed. If<WORLD_INDEX>is not in context, fall back to reading.alive/_index.yamldirectly. - If no index exists at all — generate it first (
python3 "$ALIVE_PLUGIN_ROOT/scripts/generate-index.py" "$WORLD_ROOT"), then read the output. Fall back to manual scanning only on first-time setup before the index infrastructure exists. - Freshness check — read the
generated:timestamp from the index. Display it in the dashboard header. If older than 10 minutes, show a warning. If older than 1 hour, suggest regeneration. This makes index staleness visible instead of invisible. - Build the tree from the index — parent/child relationships from
parent:field - Lightweight fresh checks — one Bash call each, no subagents, no Explore agents:
- Unsaved sessions with stash: already in the index as
unsaved_with_stash:. If non-zero, surface in the Attention section. No bash loop needed. - Unrouted inputs: resolve the world root via the doctor JSON surface, then list the absolute path. Never use a relative
ls 03_Inbox/— it silently fails when the Bash tool's cwd isn't the world root. One-liner:
Doctor's pinned-strategy result is the authoritative resolver — preferred over raw config-file reads, which mangle paths with internal whitespace. Just the filenames, no deep reads.WR=$(alive doctor --check=world-root --json | jq -r '.world_root') && [ -n "$WR" ] && ls "$WR/03_Inbox/" 2>/dev/null | grep -v '^\.' | grep -v '^Icon' - API context: only if context sources are listed in the session start injection (already in your context from the hook — do NOT re-read preferences.yaml).
- Unsaved sessions with stash: already in the index as
- Compute attention items from fresh checks + index staleness signals
- Inbox triage (background) — if
03_Inbox/has items, dispatch a background agent to triage them. Don't wait for it — render the dashboard immediately, the triage results arrive while the human reads.
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.
- 2d ago First seen · 273 lines · 63 tokens per session scan A 604d7a905a3d
alive:world is a skill published in the GitHub repository alivecontext/alive (127 stars, last pushed 7d ago), licensed MIT. It adds 63 tokens to every session and 3,222 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
crit-cli
Use when an agent needs to author or reply to crit inline comments programmatically (including multi-agent workflows commenting on shared code/plans/docs/proposals), publish or unpublish a crit review with crit share, sync a crit review to or from a GitHub PR or GitLab MR, or read/interpret a crit review JSON file.…
crit
Review code changes, a plan, a live page (running dev server), or a local HTML file with Crit inline comments and structured human feedback. Use only when the user explicitly invokes /crit or directly asks to use Crit; a generic review request does not count.
crit-story
Author a crit story only when the user explicitly invokes /crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.
cco-budget
Configure token budget limits, auto-compact settings, and view current budget status (model-aware — Opus 4.8 default, full 1M context at standard price).
cco-config
View and tune CCO's behavior thresholds — re-read warnings, cache staleness, prompt-coach length bands, and the /cco-pack budget cap.
cco-overhead
Audit the fixed context overhead every session starts with — system prompt, MCP tools, agents, CLAUDE.md, memory — measured from real transcript usage.