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 skills add Szotasz/marveen --skill channel-plugin-duplicate-socketgit clone --depth 1 https://github.com/Szotasz/marveenWrote 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/szotasz/marveen/channel-plugin-duplicate-socket)<a href="https://agentmods.dev/skills/szotasz/marveen/channel-plugin-duplicate-socket"><img src="https://agentmods.dev/badge/skills/szotasz/marveen/channel-plugin-duplicate-socket.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 | $0.00110 | $0.01551 |
| Opus 5 | $0.00055 | $0.00776 |
| Sonnet 5 | $0.00022 | $0.00310 |
| Haiku 4.5 | $0.00011 | $0.00155 |
Grade C, and why
channel-plugin-duplicate-socket scanned grade C with 2 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 8d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **MANAGED allowlist trap (Claude Code 2.1.143)**: az `allowedChannelPlugins` MANAGED setting (system-level admin policy). macOS-en `/Library/Application Support/ClaudeCode/managed-settings.json`. User-szintű `~/.claude Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
3. Megnézed a user-szintű engedélyezést: `grep -A5 enabledPlugins ~/.claude/settings.json`. Ha a problémás plugin `true` itt, akkor minden agent betölti. How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Channel plugin duplicate Socket Mode connection
Mikor használd
- Új channel plugin (slack-channel, vagy bármi Socket-Mode-os) telepítve a flottába
- Egy agent (pl. the Slack agent) használja a Slack-et, és the operator azt jelzi hogy DM-eket küld, de "semmi nem történik"
- A Slack auth.test OK, manifest OK, tokenek OK, bot scope-ok rendben — de events nem érkeznek a server.ts-hez
- Egyszerre több claude process fut
--channelsflaggel a flottában
Eljárás
ps aux | grep server.ts | grep -v grep-- count how many slack-channel server.ts processes are running- Ha >1 instance:
ps -o pid,ppid,command -p <pid>minden parent process-re. A ppid chain végén legyenclaude --continue ... --channels plugin:.... Ha az--channelsflag NEM ugyanaz mint a plugin, akkor a plugin másodlagosan loadolódott azenabledPlugins-ből. - Megnézed a user-szintű engedélyezést:
grep -A5 enabledPlugins ~/.claude/settings.json. Ha a problémás plugintrueitt, akkor minden agent betölti. - Project-szinten override-old:
<projekt>/.claude/settings.json->{"enabledPlugins": {"<plugin>@<marketplace>": false}}. Project > user precedencia. - Az adott agent saját project-szintű settings.json-jában HAGYD true-n -- ott explicit engedélyezve marad.
- A felesleges channel-process-eket killeld (kill ), vagy restart-old a tmux session-t, amelyikben fut.
- Verify:
ps aux | grep server.ts | grep -v grep-- csak az agent-szintű session-ben fusson a plugin.
Buktatók
- A Socket Mode bug NÉMA: a Slack két aktív Socket Mode connection-t load-balance-ol round-robin-szerűen, így az event-ek FELE a "nem-figyelő" server.ts-hez megy, és ott eldobódik. NEM warning, NEM error -- csak "üzenetek tűnnek el".
- A user-szintű
~/.claude/settings.json.allowedChannelPluginsésenabledPluginsKÜLÖNBÖZŐ dolgok. AzallowedChannelPluginscsak whitelistezi a--channels plugin:Xflag-et, de azenabledPlugins.X: trueaz ami minden agent-ben loadolja az MCP-t. - A flotta-architektúra MIATT (minden agent claude-code-instance) ez nem egyetlen-host probléma -- minden agent függetlenül nyit Socket-et, a Slack-side meg nem tudja megkülönböztetni őket egy workspace-en.
- Telegram-nál NEM "event eltűnés" módon jelentkezik, hanem lockfight-ként: a Telegram Bot API long-poll alapú, és Telegram lockolja a session-t (409 Conflict) ha két instance poll-ol -- a két agent felváltva kicsapja egymást. Tünet: amikor az agent-X tmux session restart-ol, az agent-Y telegram MCP "disconnected" jelez, és viszont. Megoldás: bármely agent project-szintű
.claude/settings.json-ben, amelyik NEM telegram-en kommunikál,enabledPlugins.telegram@claude-plugins-official: false-re kell tenni. A globálisunset TELEGRAM_BOT_TOKENaz agent-process.ts indítóscriptben NEM elég, mert a telegram plugin a state-dir.env-ből is olvas (a SLACK_STATE_DIR env van set-elve, de TELEGRAM_STATE_DIR nem -- így a telegram plugin a globális~/.claude/channels/telegram/.env-be esik vissza, ami az orchestrator agent tokenét tartalmazza). - Csatorna-policy külön a DM-policy-tól: a slack-channel plugin gate-jén a DM-policy (
access.dmPolicy+allowFrom) és a csatorna-policy (access.channels[channelId]) TELJESEN különálló. Egy üreschannels: {}minden csatorna-message-et (még @-mention-t is) dropp-ol. Per-csatorna opt-in:channels[C...]: {allowFrom: [U...], requireMention: true}. UX-bug forrás: setup után a user @-tageli a botot csatornában és néma marad, fél órás debug. Felhasználóbarát megoldás: SLACK_AUDIT_LOG env-be írott journal-t a dashboard polling-ozzon, gate.inbound.drop entry-kből pending_channel_requests tábla, Jóváhagy/Elutasít gomb a dashboard-on. - MANAGED allowlist trap (Claude Code 2.1.143): az
allowedChannelPluginsMANAGED setting (system-level admin policy). macOS-en/Library/Application Support/ClaudeCode/managed-settings.json. User-szintű~/.claude/settings.jsonés project-szintű.claude/settings.jsonegyik sem elég -- a tmux pane "plugin:X · not on the approved channels allowlist" warning-ot mutat. Az MCP plugin tools ettől még connected-ek (8 tools), DE a Claude Code disable-li anotifications/claude/channelinbound notification path-et a chat-be. A server.ts megkapja az event-eket (sessions/-be ír, smoke-test megerősíti), a Claude Code mégis lenyeli őket -- a pane idle marad, semmilyen<channel source="slack" ...>tag nem jelenik meg. Megoldás: sudo szükségesmkdir -p "/Library/Application Support/ClaudeCode"+ managed-settings.json létrehozása az allowedChannelPlugins-szal.
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.
- 8d ago First seen · 34 lines · 110 tokens per session scan C 2eaf40af0ee8
channel-plugin-duplicate-socket is a skill published in the GitHub repository Szotasz/marveen (98 stars, last pushed today), licensed MIT. It adds 110 tokens to every session and 1,551 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 2 findings (asks for root, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
diagnose
Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
byted-util-volcengine-detect-retry
An orchestration workflow for Volcengine Cloud Detect, a service that checks websites or network endpoints from test locations.