agent-reach
MUST USE when user wants to research/search/look up/find anything on the internet — e.g.
Skill by DealAppSeo
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/dealappseo-agent-reach
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend, writing, research, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
agent-reach- Description
- MUST USE when user wants to research/search/look up/find anything on the internet — e.g. "research this topic", "do a deep dive on X", "search the web for X", "see what people say about X", "look this up". Also MUST USE when user mentions any platform or shares any URL/link: Twitter/X, Reddit, Facebook, Instagram, YouTube, GitHub, Bilibili, XiaoHongShu, Xiaoyuzhou Podcast, LinkedIn/jobs/recruiting, V2EX, Xueqiu (stocks), RSS. 15 platforms, multi-backend routing (OpenCLI / per-platform CLIs / APIs). Zero config for 6 channels. Run `agent-reach doctor --json` to see which backend serves each platform right now. NOT for: writing reports/analysis/translation (this skill only FETCHES internet content); posting/commenting/liking (write operations); platforms that already have a dedicated skill installed (prefer that skill).
- License
- Apache-2.0
- Metadata
- homepage: https://github.com/Panniantong/Agent-Reach
- Instructions
- 192 lines in SKILL.md · view on GitHub
Before installing
- Instructions call the network
Files (9)
- LICENSE-upstream
- SKILL.md
- career.md · references/
- dev.md · references/
- finance.md · references/
- search.md · references/
- social.md · references/
- video.md · references/
- web.md · references/
SKILL.md (first lines)
<!-- GENERATED by trinity-symphony-shared/scripts/sync-agents.sh — do not edit here.
Canonical source: trinity-symphony-shared/.claude/skills/agent-reach/SKILL.md -->
<!-- TRINITY FLEET NOTE — added by us, not upstream. Everything below the next
horizontal rule is Agent Reach's own documentation, vendored verbatim from
Panniantong/Agent-Reach v1.5.0 @ 06c202b (MIT, see LICENSE-upstream). Update
it by re-vendoring, not by editing in place. -->
# Read this before you trust `agent-reach doctor`
**`doctor` answers "is the tool installed and configured", NOT "can it reach the
internet".** Those come apart, and when they do, `doctor` reports the optimistic
one.
MEASURED 2026-08-29 from a sandboxed Claude Code Remote session:
| channel | `doctor` said | an actual call did |
|---|---|---|
| YouTube (yt-dlp) | ✅ available | **FAILED** — `Tunnel connection failed: 403 Forbidden` |
| Any web page (Jina Reader) | ✅ available | **FAILED** — `curl: (56) CONNECT tunnel failed, response 403` |
| RSS/Atom (feedparser) | ✅ available | **FAILED** — `URLError ... 403 Forbidden` |
| Exa web search | configured | **NOT_CHECKED** — needs interactive OAuth; times out unattended |
| V2EX | already reported unreachable | FAILED (agrees) |
| GitHub | needs `gh` | not installed here; this fleet uses the GitHub MCP tools |
`doctor`: **3/15 available.** Actually reachable from that session: **0/15.**
To its credit Agent Reach is honest about the one it cannot settle — it refuses
to call Exa available on configuration alone, saying it did not start the remote
service to verify. That is the right shape. The gap above is the environment
blocking egress, not the tool lying.
## What that means for you, concretely
1. **Before claiming a channel works, make one real call.** Doctor is a
pre-flight, not evidence. Announce the backend you used, per rule 2 below.
2. **Read a failure by its shape.** `CONNECT tunnel failed, response 403` is the
sandbox proxy refusing — the tool is fine and the same command works on a
normal machine. An ordinary HTTP status in the response body means you
connected and the server answered. Do not report the first as a broken tool.
3. **Three outcomes, never two.** A channel you could not reach is NOT_CHECKED,
not FAILED — unless you actually got a refusal from the far end.
4. **Where this runs matters.** On a developer desktop or a T12 host with normal
egress, the zero-config channels work as upstream documents. In a sandboxed
CCR session, assume none of them do until a real call proves otherwise.
Install (machine-local, not carried by git):
```bash
git clone --depth 1 https://github.com/Panniantong/Agent-Reach.git /tmp/agent-reach
pip install /tmp/agent-reach && agent-reach doctor
```
The GitHub archive URL and PyPI are both wrong routes here: `archive/main.zip`
and `codeload.github.com` are proxy-403, and **the `agent-reach` package on PyPI
is a different project by a different author** (jgalea, v0.1.0) that happens to
share the name. Clone and install from source.
---
# Agent Reach — internet capability router
15 platforms, multiple backends each. **When this skill exists, use it for
these platforms — do not invent your own approach.**
## Standing rules (apply for the whole session)
1. **Health-check before acting**: for multi-backend/login-backed platforms (XiaoHongShu /
Reddit / Bilibili / Twitter / Facebook / Instagram), run `agent-reach doctor --json` first.
Use a populated `active_backend`; `active_backend: null` means Doctor deliberately skipped a
live probe to avoid browser-cookie reads or remote writes, not that no backend exists. Only when
the user's task requires that platform, run the reference's read-only command to verify it.
2. **Announce what you use**: say "using agent-reach, platform X via backend Y"
before starting.
3. **On failure, follow the retry chains in references/** — never guess
commands.
4. **For broad research tasks**: combine platforms (Exa for web search +
Twitter/Reddit for discussions + XiaoHongShu/Bilibili for Chinese
perspectives), collect in parallel, then synthesize.
5. **Watch versions for the user**: after finishing a substantial
multi-platform task, run `agent-reach check-update` (fast, one API call).
If a new version exists, append one line to your wrap-up: "Agent Reach
vX.Y.Z is available — paste this to me to update: 帮我更新 Agent Reach:
https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/update.md".
Never interrupt the current task to update; never nag about the same version twice.
## Routing table
| User intent | Category | Details |
|---------|------|---------|
| Web / code search | search | [references/search.md](references/search.md) |
| XiaoHongShu / Twitter / Bilibili / V2EX / Reddit / Facebook / Instagram | social | [references/social.md](references/social.md) |
| Jobs / LinkedIn | career | [references/career.md](references/career.md) |
| GitHub / code | dev | [references/dev.md](references/dev.md) |
| Web pages / articles / RSS | web | [references/web.md](references/web.md) |
| YouTube / Bilibili / podcast transcripts | video | [references/video.md](references/video.md) |
| Xueqiu / stock quotes | finance | [references/finance.md](references/finance.md) |
## Zero-config quick commands
```bash
# Exa web search
mcporter call exa.web_search_exa query="query" numResults=5
# Read any web page
curl -s "https://r.jina.ai/URL"
# GitHub search
gh search repos "query" --sort stars --limit 10
# YouTube subtitles (never use yt-dlp for Bilibili; retry chain in video.md)
yt-dlp --write-sub --write-auto-sub --skip-download -o "/tmp/%(id)s" "URL"
# V2EX hot topics
curl -s "https://www.v2ex.com/api/topics/hot.json" -H "User-Agent: agent-reach/1.0"
# Bilibili search (bili-cli, no login needed)
bili search "query" --type video -n 5
```From DealAppSeo/trustshell, Apache-2.0.
More from DealAppSeo/trustshell
- 0.565score
trustshell
DealAppSeo
Fact-check a claim, read an agent's live RepID, verify its ZK proof, or print after-create and Honesty A by shelling out to the published bins.
SkillAFreshness: AClaude Code +9 - 0.564score
cmo
DealAppSeo
Use when drafting social/edutainment units, Hard-Knocks style shorts, or TrustHarness marketing.
SkillAFreshness: AClaude Code +9 - 0.519score
verify-first
DealAppSeo
Before claiming what exists, which service has a secret, what T12 is, or what a command does — open the file, the live CLI help, or the API.
SkillBFreshness: BClaude Code +9 - 0.500score
impeccable
DealAppSeo
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or…
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 6
- Last commit (this folder)
- Aug 29, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/DealAppSeo/trustshell
- Website
- trustshell.dev
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
More from DealAppSeo
- 0.565score
trustshell
DealAppSeo
Fact-check a claim, read an agent's live RepID, verify its ZK proof, or print after-create and Honesty A by shelling out to the published bins.
SkillAFreshness: AClaude Code +9 - 0.564score
cmo
DealAppSeo
Use when drafting social/edutainment units, Hard-Knocks style shorts, or TrustHarness marketing.
SkillAFreshness: AClaude Code +9 - 0.519score
verify-first
DealAppSeo
Before claiming what exists, which service has a secret, what T12 is, or what a command does — open the file, the live CLI help, or the API.
SkillBFreshness: BClaude Code +9 - 0.500score
impeccable
DealAppSeo
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or…
SkillBFreshness: BClaude Code +9
Cite this page
agent-reach skill. Agent Mart. https://agentmart.dev/skills/dealappseo-agent-reach (last updated 2026-10-24)