oliverv/caro / .agents/skills/oma-search
oma-search
VerifiedFind external documentation, web sources, or remote code with citations. Local code navigation uses the configured code tools.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.8 KB |
| resources/checklist.md | markdown | 1.7 KB |
| resources/error-playbook.md | markdown | 5.1 KB |
| resources/examples.md | markdown | 4.2 KB |
| resources/execution-protocol.md | markdown | 3.0 KB |
| resources/intent-rules.md | markdown | 2.7 KB |
| resources/trust-registry.md | markdown | 5.0 KB |
resources/trust-registry.md
View on GitHub# Trust Registry Reference
Domain trust scoring strategy for search results.
**Single source of truth: the CLI.** Resolve every score by running:
```bash
oma search trust <domain>
# → { "domain": "...", "level": "verified|community|external|unknown",
# "score": 0.95, "tags": ["..."], "source": "registry|heuristic|tranco" }
```
The registry, heuristics, and Tranco fallback are implemented in
`cli/commands/search/trust.ts`. Do NOT hand-score domains or call ranking
APIs directly; this document describes what the CLI returns so you can
interpret it.
## Trust Levels
| Level | Score Range | Label | Description |
|-------|-----------|-------|-------------|
| verified | 0.85-0.95 | **** | Official documentation, vendor sites, standards bodies |
| community | 0.50-0.75 | *** | User-generated, curated platforms |
| external | 0.20-0.49 | ** | Third-party content sources |
| unknown | (none) | (none) | Cannot determine trust level |
## Scoring Rules
1. **Domain-level only**: score applies to the entire domain, not sub-paths
2. **Unknown domains are NOT excluded**: they appear with `—` label
3. **`--strict` filter**: only shows results with score >= 0.85 (verified+)
4. **Sort tiebreaker**: when relevance is equal, higher trust score ranks first
## How the CLI Resolves a Score
Resolution order inside `oma search trust` (first hit wins):
### 1. Static registry (`source: "registry"`)
Enumerated domains with fixed scores. Representative entries
(see `trust.ts` REGISTRY for the full list):
| Domain | Level | Score | Tags |
|--------|-------|-------|------|
| `github.com`, `gitlab.com` | verified | 0.95 | code-host |
| `docs.github.com`, `developer.mozilla.org`, `typescriptlang.org`, `react.dev`, `go.dev`, `rust-lang.org`, `python.org`, `nodejs.org` | verified | 0.95 | lang-docs |
| `w3.org`, `tc39.es`, `ietf.org`, `datatracker.ietf.org`, `owasp.org` | verified | 0.95 | standards |
| `doi.org` | verified | 0.95 | academic |
| `nextjs.org`, `vercel.com` | verified | 0.90 | vendor |
| `npmjs.com`, `pypi.org`, `crates.io`, `pub.dev` | verified | 0.90 | registry |
| `arxiv.org` | verified | 0.90 | academic |
| `wikipedia.org` | community | 0.75 | encyclopedia |
| `stackoverflow.com`, `stackexchange.com` | community | 0.70 | qna |
| `news.ycombinator.com` | community | 0.65 | news |
| `reddit.com` | community | 0.55 | forum |
| `freecodecamp.org`, `baeldung.com` | external | 0.45 | tutorial |
| `dev.to` | external | 0.40 | blog |
| `medium.com`, `hashnode.com`, `substack.com` | external | 0.35 | blog |
| `velog.io`, `tistory.com` | external | 0.30 | blog, kr |
| `w3schools.com`, `geeksforgeeks.org` | external | 0.30 | tutorial |
### 2. Heuristic patterns (`source: "heuristic"`)
| Pattern | Level | Score | Rationale |
|---------|-------|-------|-----------|
| TLD `*.gov`, `*.edu`, `*.mil` | verified | 0.90 | Institutional domains |
| `*.gov.xx`, `*.ac.xx` (country codes) | verified | 0.85 | Institutional domains |
| `docs.*` or `*.docs.*` subdomain | verified | 0.90 | Official documentation subdomain |
| `developer.*` / `developers.*` subdomain | verified | 0.85 | Developer portal convention |
### 3. Tranco popularity prior (`source: "tranco"`)
For domains missing both above, the CLI queries the Tranco list
(`tranco-list.eu` API, SSRF-guarded) and maps rank to a conservative score:
| Tranco Rank | Level | Score |
|-------------|-------|-------|
| < 10,000 | community | 0.60 |
| < 100,000 | external | 0.40 |
| >= 100,000 | external | 0.20 |
| Not ranked / API failure | unknown | `—` |
**Popularity is a prior, not authority**: the Tranco path is capped at 0.60,
so a merely-popular domain can never reach `verified` or pass `--strict`.
## Agent-Level Rules (outside the CLI)
Two cases the CLI cannot decide are handled by the agent:
1. **Context7-resolved docs**: results returned by the `docs` route come from
Context7 library resolution — label them `verified 0.95` (tag `lib-docs`)
without a CLI lookup; resolution IS the verification.
2. **Official-site upgrade (upgrade-only)**: when a result domain is
self-evidently the official site of the queried library or vendor
(e.g. `htmx.org` for an htmx query, `redis.io` for Redis), the agent may
assign `verified 0.90` with tag `official-site` even if the CLI returns a
lower score. Downgrades are never allowed; every other domain uses the CLI
score as-is.
## Caching
Cache resolved scores in memory for the current session to avoid repeated CLI calls. No code-intelligence memory provider is required. If persistence is needed, use a generated path under `.agents/state/`, record the resolution time and source, and refresh it before relying on it in a later session.
## Lookup Algorithm
```
1. Extract domain from result URL (strip protocol, path, query)
2. Check the current session’s in-memory domain cache
3. If cache miss → run `oma search trust <domain>`
4. Apply agent-level rules (Context7 docs label, official-site upgrade)
5. Attach [level, tags, score] to result; unknown → label `—`, keep result
6. Write newly resolved scores to cache
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |