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 cometchat/cometchat-skills --skill cometchat-react-v7-featuresgit clone --depth 1 https://github.com/cometchat/cometchat-skillsWrote 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/cometchat/cometchat-skills/cometchat-react-v7-features)<a href="https://agentmods.dev/skills/cometchat/cometchat-skills/cometchat-react-v7-features"><img src="https://agentmods.dev/badge/skills/cometchat/cometchat-skills/cometchat-react-v7-features/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cometchat/cometchat-skills/cometchat-react-v7-features"><img src="https://agentmods.dev/badge/skills/cometchat/cometchat-skills/cometchat-react-v7-features.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00092 | $0.03292 |
| Opus 5.5 | $0.00037 | $0.01317 |
| Sonnet 5.5 | $0.00018 | $0.00658 |
| Haiku 4.5 | $0.00009 | $0.00329 |
Grade A, and why
cometchat-react-v7-features 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 1mo 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ground truth: the feature existence oracle is
features.json(pack root). A feature a user asks for MUST be anidthere — if it's not listed, it does not exist; do not invent it. Exact enablement steps + current dashboard nav + config keys: FETCH from the feature'sdocs_topic(docs flag dashboard paths as fragile). UI component names come from theweb-v7catalog. Status: oracle curated from live docs 2026-07-29; entries marked VERIFY need confirming before shipping enablement steps.
Companion skills (read first)
cometchat-react-v7-core— install, credentials,init→login→render, provider. This skill ASSUMES it.cometchat-react-v7-components— the component catalog features render through.
Use this skill when
The user wants to turn ON a specific capability: "add reactions", "enable polls", "turn on smart replies", "add push notifications", "enable message translation", "add calling", "add moderation".
Prerequisites & install
Core setup done (credentials + init/login). Most features add no package; voice-video-calls needs the separate calls package.
Enable a feature (BODY — the oracle-driven flow)
- Look it up in
features.json. Match the request to a featureid. Not listed → tell the user it isn't an available feature (don't invent one). - Branch on
category+enablement:default/auto→ the SDK-side capability is on, but checkneeds_stitchingfirst. Ifneeds_stitching: true, the drop-in renders a default-ON affordance that is INERT/partial until the host wires it — you MUST either stitch it (callback + companion panel/component) or hide it; never leave it dead-ending (see the "Default-on affordances" pitfall). Otherwise nothing to enable — point at theui_componentsif they want to surface/customize it.extension+dashboard-extension→ enable the extension, then wire the client. Two ways to enable: (a) via the CLI —npx @cometchat/skills-cli@3 features enable <id>flips the dashboard toggle via the API (reads the App ID from.cometchat/config.json;features listshows current state); OR (b) manually in the CometChat Dashboard → Extensions. Offer the CLI (faster), fall back to the dashboard. The CLI only flips the toggle — YOU still wire the client component/prop. FETCH the current dashboard steps fromdocs_topic— do not bake the path.ai→ enable via dashboard-toggle (AI User Copilot) AND wire the code prop — NOT zero-code (AUDIT-009). The toggle can be flipped by the CLI too —@cometchat/skills-cli@3 features enable <id>(an AI feature may first need an OpenAI key set on the app; the CLI reports that). Then wire the props: The copilot props DEFAULT TO OFF, so nothing appears until you set them: smart-reply →showSmartRepliesonCometChatMessageList(default false; renders in the list FOOTER, not the composer); conversation-starter →showConversationStartersonCometChatMessageList(default false); conversation-summary →showConversationSummaryButton+onSummaryClickonCometChatMessageHeader(default false). These areneeds_stitching:true. Onlyai-assistantrenders a distinct component (CometChatAIAssistantChat). Plan-gating is often 'unknown' — verify, don't assert.package-install(calls) → install the separate calls package (FETCH the exact coordinate/version fromdocs_topic), then wire theui_components.- No UI Kit component (SDK-backed feature) → the capability exists in the SDK but the UI Kit ships NO component for it (AI agents, campaigns, advanced/AI moderation, bots, webhooks, transient messages). Wire it from the JavaScript Chat SDK directly — look the exact method up in
cometchat-react-v7-core/references/docs-map.md→ SDK docs (thellms.txtindex → the feature's page → its AI-Integration-Quick-Reference accordion for the method signatures; e.g. AI agents →CometChat.addAIAssistantListener(...)). This is the UI-Kit-first / SDK-fallback rule (RULES.md) — don't skip the feature and don't hand-roll what the SDK already exposes. dashboard-only(moderation) → configured ENTIRELY in the dashboard; emit ZERO moderation UI code (AUDIT-078). Itsui_components(CometChatModerationView/CometChatFlagMessageDialog) are AUTO-RENDERED / INTERNAL, NOT host-wired: the kit auto-shows the moderation footer for blocked/pending messages (CometChatMessageList hideModerationViewdefaults false = shown), and the message list already ships a built-in "Report message" option that opens the flag dialog internally. So do NOT hand-roll a custom Report option (plugingetOptions) and do NOT mountCometChatFlagMessageDialogyourself — that duplicates the built-in. The only work is in the Dashboard: set up the Rules Engine (auto block/flag on send) + turn ON Report Message / Report User; the kit then surfaces everything with no code. (Rules Engine is not a classic extension → the CLI can't toggle it; give the user the dashboard steps.)
- Wire the UI using only the feature's
ui_components(they exist in theweb-v7catalog); fetch exact props from each component's.mdtwin (cometchat-react-v7-core/references/docs-map.md). If the feature is a whole TASK with an end-to-end guide (check the docs-map "Task guides (recipes)" section — e.g. group chat, search, threads, new-chat, message-privately, block/unblock), BUILD FROM the guide and COMPARE your implementation against it for completeness — then apply the hardening deltas on top (docs-maximal; docs-map). No matching guide? → wire theui_componentsthe normal way (above) + fetch props; the guide is an enhancement, not required. - Respect VERIFY notes — any entry noting VERIFY (or an
ai/e2e-encryption/moderationentry) must be confirmed against live docs for plan-gating and exact slugs before you emit steps.
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.
- 1mo ago First seen · 93 lines · 92 tokens per session scan A 8046600a244f
cometchat-react-v7-features is a skill published in the GitHub repository cometchat/cometchat-skills (162 stars, last pushed 22d ago), licensed MIT. It adds 92 tokens to every session and 3,292 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-12.
cometchat-react-v7-features compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| cometchat-react-v7-featurescometchat/cometchat-skills | Skill | 162 | 22d ago | MIT | A | 92 | Claude Code |
| vercel-react-best-practicescalesthio/OpenMontage | Skill | 77k | 3d ago | AGPL-3.0 | A | 67 | Claude Code, Codex |
| vercel-react-native-skillsfcakyon/claude-codex-settings | Skill | 1.2k | yesterday | Apache-2.0 | A | 57 | Claude Code |
| vercel-composition-patternscalesthio/OpenMontage | Skill | 77k | 3d ago | AGPL-3.0 | A | 58 | Claude Code, Codex |
| frontendcode-yeongyu/oh-my-openagent | Skill | 71k | today | none | A | 79 | Claude Code, Codex |
| Error Boundary TesterPramodDutta/qaskills | Skill | 245 | 3d ago | MIT | A | 26 | Claude Code, Codex |
| 21st-ui21st-dev/magic-mcp | Skill | 6.4k | 1mo ago | ISC | A | 63 | Claude Code |