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/search-worldnpx skills add alivecontext/alive --skill search-worldgit 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/search-world)<a href="https://agentmods.dev/skills/alivecontext/alive/search-world"><img src="https://agentmods.dev/badge/skills/alivecontext/alive/search-world.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 5.1 | $0.00064 | $0.00965 |
| Opus 5 | $0.00032 | $0.00483 |
| Sonnet 5 | $0.00013 | $0.00193 |
| Haiku 4.5 | $0.00006 | $0.00097 |
Grade A, and why
alive:search-world scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Find
Search across the world. One verb for all retrieval.
How It Searches
Priority order — fastest and highest signal first:
1. Frontmatter Scan (fast, structured)
Scan _kernel/key.md across all walnuts. Matches on: type, goal, people names, tags, links, reference descriptions.
2. Insights Search (standing knowledge)
Scan _kernel/insights.md across relevant walnuts. Domain knowledge that persists — "Nova Station test windows are Tue-Thu only."
3. Log Search (decisions, history)
Search _kernel/log.md entries. Signed decisions, session summaries, what happened when. Frontmatter first (last-entry, summary), then entry bodies.
4. Task Search (work queue)
Search tasks across all walnuts using tasks.py list with the --world and --search flags:
python3 "$ALIVE_PLUGIN_ROOT/scripts/tasks.py" list --world "$WORLD_ROOT" --search "{query}"
Returns JSON with tasks matching the query across all walnuts. Each result includes a walnut field for attribution. Combine with --status, --priority, --assignee, --tag for narrower results.
5. Working File Search (drafts)
Scan */ across walnuts (bundles are flat in walnut root). Find drafts by name, version, age, squirrel attribution.
6. Bundle Manifest Search (captured content metadata)
Search */context.manifest.yaml files across walnuts (bundles are flat in walnut root). Match on frontmatter: type, date, source, participants, subject.
7. Raw Reference Search (expensive)
Load actual raw files. Only on explicit request — "read me that email from Jax."
8. Context Source Cascade (if nothing found locally)
If steps 1-7 return nothing and the user believes it exists, automatically fan out to configured context sources before reporting "not found." Check .alive/preferences.yaml context_sources: for what's available — any configured MCP server, sync script, or API connection is a valid search target.
One-hop inference applies: if the user says "the setup guide I sent Sarah" and an email integration is configured, that's enough to trigger a search without being asked. The system should resolve across its full context surface — local files are not the only source of truth.
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.
- 5d ago First seen · 103 lines · 64 tokens per session scan A e43737e29c3a
alive:search-world is a skill published in the GitHub repository alivecontext/alive (127 stars, last pushed 10d ago), licensed MIT. It adds 64 tokens to every session and 965 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-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.
cco-pack
Build an optimal context pack for the user's task — ranked file list with offset/limit suggestions, based on git state, mentioned paths, and historical patterns.
cco-templates
Manage context templates for common task types.
cco
Context Control Center — one screen for budget, $ spent, tokens saved, waste, last prompt grade, the active task, and ready-to-run optimization actions.
cco-doctor
Health check for the context-optimizer plugin install — verifies versions, hooks, data dir, model config, and reports any issues.
cco-replay
Show recent session summaries for quick context recovery.