What is cometchat-audit?

A read-only review of an existing CometChat integration. It checks security, setup order, version mismatches, release readiness, accessibility, and localization, meaning support for different languages and regions.

Use it to inspect credentials, server-token use, initialization and login flow, rendered components, version conflicts, unused interface actions, accessibility gaps, and localization issues.

It gives you a ranked list of risks and fixes before a launch, security review, or upgrade, rather than changing code without approval. It also stops when no CometChat integration can be detected.

cometchat-audit — Skill for Claude Code by cometchat/cometchat-skills

Install

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.

Any agent

npx skills add cometchat/cometchat-skills --skill cometchat-audit

Clone the repo

git clone --depth 1 https://github.com/cometchat/cometchat-skills

Made for: Claude Code.

Or install cometchat-skills, the plugin that ships this one along with the rest of its 98 skills, 1 MCP server.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/cometchat/cometchat-skills/cometchat-audit/github.svg)](https://agentmods.dev/skills/cometchat/cometchat-skills/cometchat-audit)

Your own site

<a href="https://agentmods.dev/skills/cometchat/cometchat-skills/cometchat-audit"><img src="https://agentmods.dev/badge/skills/cometchat/cometchat-skills/cometchat-audit/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/cometchat/cometchat-skills/cometchat-audit"><img src="https://agentmods.dev/badge/skills/cometchat/cometchat-skills/cometchat-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 137 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,340 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00137	$0.01340
Opus 5.5	$0.00055	$0.00536
Sonnet 5.5	$0.00027	$0.00268
Haiku 4.5	$0.00014	$0.00134

Measured 1mo ago against content hash c462666da6fc, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

cometchat-audit 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.

skills/cometchat-audit/SKILL.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ground truth: the authority for "correct" is RULES.md + the resolved family's skills (cometchat-<family>-core/-production/-troubleshooting) and the offline probe npx @cometchat/skills detect --json. Verify every symbol/prop you flag against the family catalog + docs (via -core/references/docs-map.md); never flag from memory. This is a read-only review — produce the report first; only change code if the user says so (then hand each fix to the owning skill).

Use this skill when

Someone wants an existing CometChat integration checked — before a launch, a security review, an upgrade, or a "why is this flaky / is this safe" question. It is the proactive complement to the per-family troubleshooting skills (cometchat-<family>-troubleshooting, which react to a specific symptom).

How to run the audit

Detect — npx @cometchat/skills detect --json: framework, installed UI Kit, version_conflict, existing_cometchat. Resolve <family> from peers.yaml. If nothing is detected, say so and stop (nothing to audit).
Read the integration — the init/login/lifecycle file, where the surface renders, the env/secret files, the server token endpoint (if any), and any custom message/theme/feature code. Do NOT read node_modules/.d.ts.
Score each check below, gather evidence (file:line), and write a ranked report — most severe first — with, per finding, the fix and the skill that owns it. Then offer to apply fixes; don't auto-edit.

The checklist

Security (highest severity)

Auth Key present in client code / bundle / binary? → critical; must move to server-minted tokens (cometchat-security, cometchat-<family>-production).
Token endpoint deriving the UID from a client parameter (?uid=)? → impersonation.
REST API Key anywhere client-side or in the repo? → critical.
Sessions revocable on offboarding (flush tokens)? RBAC roles configured, or everyone on default? (cometchat-security.)

Correctness

init() → login() → render order honored; no component rendered before login resolves; no init during SSR; StrictMode/double-invoke guarded.
initFromSettings used (ai-agent attribution), not the classic builder init.
Surface has real dimensions (not a collapsed 0-height box); host CSS not leaking into .cometchat.
No raw localization keys rendered (group_info etc.).
No dead affordances — every default-on control wired or hidden.
Listeners removed on unmount/dispose (no duplicate messages/leaks).

Read the full file on GitHub · 62 lines

Changes

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 · 62 lines · 137 tokens per session scan A c462666da6fc

Subscribe to this mod's changes

cometchat-audit is a skill published in the GitHub repository cometchat/cometchat-skills (162 stars, last pushed 22d ago), licensed MIT. It adds 137 tokens to every session and 1,340 once invoked, about $0.0005 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 2026-09-24.

Claude Code Code review Security agent-skills ai-agent chat cometchat cursor

cometchat-audit next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
cometchat-auditcometchat/cometchat-skills	Skill	162	22d ago	MIT	A	137	Claude Code
codebase-designPrismer-AI/PrismerCloud	Skill	1.4k	15d ago	MIT	A	37	Claude Code, Codex
review-workcode-yeongyu/oh-my-openagent	Skill	71k	yesterday	none	A	63	Claude Code, Codex
tech-lead/code-reviewechoVic/boss-skill	Skill	547	5d ago	MIT	A	52	Claude Code
pr-blocker-summarizerFrancyJGLisboa/agent-skill-creator	Skill	2.4k	1mo ago	MIT	A	54	Claude Code
skill-trigger-auditorAlexZio00/sovereign-skills	Skill	151	3d ago	MIT	A	38	Claude Code, Codex
ponytail-lazy-senior-devGulajavaMinistudio/awesome-copilot-id	Skill	93	9d ago	MIT	A	146	Claude Code