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/demonpx skills add alivecontext/alive --skill demogit 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/demo)<a href="https://agentmods.dev/skills/alivecontext/alive/demo"><img src="https://agentmods.dev/badge/skills/alivecontext/alive/demo.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 | $0.00052 | $0.10344 |
| Opus 5 | $0.00026 | $0.05172 |
| Sonnet 5 | $0.00010 | $0.02069 |
| Haiku 4.5 | $0.00005 | $0.01034 |
Grade A, and why
alive:demo 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 — 771 lines — stays where its author put it; the contents beside it link to each section on GitHub.
alive:demo -- generative demo-world skill
This is the router. Route on $ARGUMENTS[0] to a sibling .md file (when present) or to the interactive create flow. Six subcommands + an interactive default:
$ARGUMENTS[0] |
Routes to | Status |
|---|---|---|
empty / create |
Interactive: preset vs. custom; create.md for the custom path |
wired |
list |
list.md |
wired |
activate <ref> |
activate.md |
wired |
deactivate |
deactivate.md |
wired |
delete <ref> |
delete.md |
wired |
status |
status.md |
wired |
reset |
rebuilds demo-state.json (recovery from schema mismatch) | wired |
spike-test |
retired (fn-2-2zz.1 spike) | removed below |
All five non-create sibling .md files (list.md, activate.md, deactivate.md, delete.md, status.md) invoke their CLI subcommand under $ALIVE_PLUGIN_ROOT/bin/alive demo <name> and print the CLI's rendered_block field verbatim. create.md is the custom-path orchestrator: it walks Stage 0 -> Stage 1 hand-off -> Stage 2/3/4/5 driver chain end-to-end. The interactive default below routes the human into either the preset path or create.md based on their pick.
The router emits bordered blocks INLINE as markdown — matches the convention in plugins/alive/skills/world/SKILL.md and plugins/alive/skills/save/SKILL.md. There is no Python render_block bridge: the LLM emits its own markdown; only the CLI uses lib.format_block (per the codex review which rejected forcing LLM formatting from Python).
User choices use AskUserQuestion, never inline 1. / 2. / 3. answer text. Numbered options inside a bordered block are advisory; the actual decision flows through the question tool.
State + paths (canonical)
| Concern | Path | Override? |
|---|---|---|
| Demo state file | ~/.config/alive/demo-state.json |
NO. Tests set HOME to tmp_path. |
| Demo state lock | ~/.config/alive/.demo-state.lock |
NO. |
| World-root pointer | ~/.config/alive/world-root |
NO — owned by _world_root_io.write_world_root_file (#64). |
| Demo worlds + partials | ~/.alive-demos/ |
YES via $ALIVE_DEMO_BASE_DIR. |
What ships with it
42 files 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.
- activate.md 6.8 KB
- anchor_confirm.md 11 KB
- cli_register.py 106 KB runs code
- create.md 17 KB
- deactivate.md 2.7 KB
- delete.md 3.5 KB
- lib.py 60 KB runs code
- list.md 3.5 KB
- preset/realistic-seeded/_world_meta.json 1.3 KB
- preset/realistic-seeded/02_Life/people/jax-stellara/_kernel/insights.md 308 B
- preset/realistic-seeded/02_Life/people/jax-stellara/_kernel/key.md 900 B
- preset/realistic-seeded/02_Life/people/jax-stellara/_kernel/log.md 845 B
- preset/realistic-seeded/02_Life/people/ryn-okata/_kernel/insights.md 627 B
- preset/realistic-seeded/02_Life/people/ryn-okata/_kernel/key.md 968 B
- preset/realistic-seeded/02_Life/people/ryn-okata/_kernel/log.md 1.2 KB
- preset/realistic-seeded/04_Ventures/nova-station/_kernel/completed.json 1.3 KB
- preset/realistic-seeded/04_Ventures/nova-station/_kernel/insights.md 1.7 KB
- preset/realistic-seeded/04_Ventures/nova-station/_kernel/key.md 1.5 KB
- preset/realistic-seeded/04_Ventures/nova-station/_kernel/log.md 4.2 KB
- preset/realistic-seeded/04_Ventures/nova-station/_kernel/tasks.json 590 B
- preset/realistic-seeded/04_Ventures/nova-station/launch-readiness/context.manifest.yaml 979 B
- preset/realistic-seeded/04_Ventures/nova-station/launch-readiness/tasks.json 590 B
- preset/realistic-seeded/04_Ventures/nova-station/shielding-review/context.manifest.yaml 875 B
- preset/realistic-seeded/04_Ventures/nova-station/shielding-review/tasks.json 18 B
- preset/realistic-seeded/README.md 4.3 KB
- README.md 11 KB
- resume.md 4.8 KB
- scaffold.py 74 KB runs code
- stages/__init__.py 996 B runs code
- stages/activate_existing.py 18 KB runs code
- stages/deactivate.py 5.6 KB runs code
- stages/delete_existing.py 8.8 KB runs code
- stages/preset.py 34 KB runs code
- stages/stage0.py 57 KB runs code
- stages/stage1.py 41 KB runs code
- stages/stage2.py 81 KB runs code
- stages/stage3.py 50 KB runs code
- stages/stage4.py 53 KB runs code
- stages/stage5.py 6.7 KB runs code
- state.py 45 KB runs code
- status.md 2.3 KB
- validate.py 49 KB runs code
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 · 771 lines · 52 tokens per session scan A 759e4ed2ddbe
alive:demo is a skill published in the GitHub repository alivecontext/alive (127 stars, last pushed 8d ago), licensed MIT. It adds 52 tokens to every session and 10,344 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
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-budget
Configure token budget limits, auto-compact settings, and view current budget status (model-aware — Claude 5 lineup, Opus 5 default fallback, full 1M context at standard price).
cco-overhead
Audit the fixed context overhead every session starts with — system prompt, MCP tools, agents, CLAUDE.md, memory — measured from real transcript usage.
cco-patterns
Share learned file patterns across a team — export an anonymized digest of what's usually waste/useful/co-edited, and import a teammate's so a fresh clone benefits day one.
smart-loader
Automatically suggests optimal files to preload based on the user's task description and historical context patterns. Activates when the user starts a new task, mentions reading files, or when session context is being set up.
cco-coach
Analyze the user's last prompt for clarity, scope and specificity — give a quality score and concrete suggestions to make the next prompt produce better results.