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 happy520ai/unified-ai-system --skill unified-ai-gatewaygit clone --depth 1 https://github.com/happy520ai/unified-ai-systemWrote 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/happy520ai/unified-ai-system/unified-ai-gateway)<a href="https://agentmods.dev/skills/happy520ai/unified-ai-system/unified-ai-gateway"><img src="https://agentmods.dev/badge/skills/happy520ai/unified-ai-system/unified-ai-gateway/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/happy520ai/unified-ai-system/unified-ai-gateway"><img src="https://agentmods.dev/badge/skills/happy520ai/unified-ai-system/unified-ai-gateway.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 79 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- medium MCP Rug Pull · line 57 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium MCP Rug Pull · line 61 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium MCP Rug Pull · line 126 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
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 | $0.00035 | $0.03202 |
| Opus 5 | $0.00017 | $0.01601 |
| Sonnet 5 | $0.00007 | $0.00640 |
| Haiku 4.5 | $0.00003 | $0.00320 |
Grade A, and why
unified-ai-gateway scanned grade A with 1 finding 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 12d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
grep -RInE 'child_process|spawn\(|fetch\(|AI_GATEWAY_MCP_URL|process\.env|writeFile|appendFile|unlink|rm\(' "$REVIEW_DIR/rootfs/app/packages/mcp-server/src" "$REVIEW_DIR/rootfs/app/packages/shared-sdk/src" > "$REVIEW_DIR How it starts
The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Unified AI Gateway
Overview
Use the official unified-ai-system MCP server to inspect and exercise a local
AI gateway without provider credentials. This skill file provides operating
guidance; it does not install the server or change Codex configuration by
itself. The official Codex plugin bundles the MCP definition, while skill-only
installations require the manual setup below.
Version Note
The current public project release and latest reviewed immutable MCP image are
both v0.4.9. The inspection procedure below pins its recorded digests; those
values must not be silently replaced with a mutable tag. For a normal
provider-free demo, use the current v0.4.9 command in the
project README.
A new content review is required before changing this pinned procedure.
Prerequisites And Setup
- Confirm that Codex CLI and Docker are installed and Docker is running.
- If the 12 tools are already visible, skip setup and do not register a duplicate server.
- Explain the first stage: it downloads one reviewed platform from the
immutable
0.4.9multi-platform index into Docker's cache, inspects its metadata and layer history, creates but never starts a temporary container, exports its root filesystem, removes that temporary container, and writes an inspection inventory to a temporary directory. The reviewed platforms are linux/amd64 and linux/arm64. Obtain explicit user approval for those download and inspection changes only. - After that first approval, pull the reviewed platform manifest and complete the inspection. Do not execute the image or register it yet:
IMAGE='ghcr.io/happy520ai/unified-ai-system/mcp-server@sha256:751a0d32acd2d6b1da6ad9ac67987fbd1ff36ce26b7160014d8605f18b7907b3'
PLATFORM='linux/amd64' # Use linux/arm64 only on a reviewed ARM64 engine.
REVIEW_DIR="$(mktemp -d)"
docker pull --platform "$PLATFORM" "$IMAGE"
docker image inspect "$IMAGE" --format 'Id={{.Id}} OS={{.Os}} Architecture={{.Architecture}} User={{json .Config.User}} Entrypoint={{json .Config.Entrypoint}} Cmd={{json .Config.Cmd}} Labels={{json .Config.Labels}}'
docker image history --no-trunc "$IMAGE" > "$REVIEW_DIR/image-history.txt"
REVIEW_CONTAINER="$(docker create --platform "$PLATFORM" --pull never --entrypoint /bin/true "$IMAGE")"
docker export --output "$REVIEW_DIR/rootfs.tar" "$REVIEW_CONTAINER"
docker rm "$REVIEW_CONTAINER"
tar -tf "$REVIEW_DIR/rootfs.tar" > "$REVIEW_DIR/rootfs-files.txt"
mkdir -p "$REVIEW_DIR/rootfs"
tar --same-permissions -xf "$REVIEW_DIR/rootfs.tar" -C "$REVIEW_DIR/rootfs"
find "$REVIEW_DIR/rootfs/app" -type f -print > "$REVIEW_DIR/app-files.txt"
: > "$REVIEW_DIR/app-links.txt"
while IFS= read -r -d '' APP_LINK; do
ls -ld -- "$APP_LINK" >> "$REVIEW_DIR/app-links.txt"
done < <(find "$REVIEW_DIR/rootfs/app" \( -type l -o -type f -links +1 \) -print0)
: > "$REVIEW_DIR/native-binaries.sha256"
while IFS= read -r -d '' NATIVE_BINARY; do
sha256sum -- "$NATIVE_BINARY" >> "$REVIEW_DIR/native-binaries.sha256"
done < <(find "$REVIEW_DIR/rootfs/app" -type f -name '*.node' -print0)
find "$REVIEW_DIR/rootfs" -type f \( -perm -0100 -o -perm -0010 -o -perm -0001 \) -print > "$REVIEW_DIR/executable-files.txt"
find "$REVIEW_DIR/rootfs" -type f \( -perm -4000 -o -perm -2000 \) -print > "$REVIEW_DIR/suid-sgid-files.txt"
find "$REVIEW_DIR/rootfs/app" -type f \( -name '.env' -o -name '.env.*' -o -name '*.pem' -o -name '*.key' -o -name '*.p12' -o -name '*.pfx' -o -path '*/.ssh/id_*' \) -print > "$REVIEW_DIR/credential-like-files.txt"
grep -RInHE --include='package.json' '"(preinstall|install|postinstall|prepare|prepack|postpack)"' "$REVIEW_DIR/rootfs/app" > "$REVIEW_DIR/lifecycle-hooks.txt"
grep -RInE 'child_process|spawn\(|fetch\(|AI_GATEWAY_MCP_URL|process\.env|writeFile|appendFile|unlink|rm\(' "$REVIEW_DIR/rootfs/app/packages/mcp-server/src" "$REVIEW_DIR/rootfs/app/packages/shared-sdk/src" > "$REVIEW_DIR/runtime-sensitive-code.txt"
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 241 lines · 35 tokens per session scan A 278167281288
unified-ai-gateway is a skill published in the GitHub repository happy520ai/unified-ai-system (6 stars, last pushed today), licensed Apache-2.0. It adds 35 tokens to every session and 3,202 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
omni-compression
Configure RTK (command output), Caveman (prose), and stacked compression modes. Manage language packs, custom rules, and test prompt compression reducing tokens by 60–90%.
cli-compression
Configure and test prompt compression from the CLI. Manage RTK filters, Caveman rules, stacked compression modes, and preview compression output with real prompts.
n8n-agents
Design n8n AI agents the right way. Use when building or editing any @n8n/n8n-nodes-langchain. AI node — an AI Agent, LLM chain, Text Classifier, or Information Extractor — and whenever the user mentions AI agents, LLM with tools, tool calling, $fromAI, system prompts, agent memory, sessionId, structured/JSON output…
keirouter-chat
Chat / code generation via KeiRouter using OpenAI /v1/chat/completions or Anthropic /v1/messages format with streaming + auto-fallback combos. Use when the user wants to ask an LLM, generate code, summarize text, or run prompts through KeiRouter.
coding-agents-prompt-authoring
To author, adapt, review, and validate prompts (skills, agents, workflows, rules, etc.) with brief, contracts, and a validation pack.
playground
Author, edit, or iterate on prompts in the Phoenix prompt playground, including running experiments over a dataset. Load before any playground ui. operation call, including single-shot prompt rewrites.