Decide and implement how to support a new LLM provider or agent engine in AWF - either as a proxied provider (api-proxy adapter) or a direct-API engine (domain allowlist only), e.g.
OfficialMITAuto-check passed
Install Add LLM Provider
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a claude-code
Project install by default; add -g for ~/.claude/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .claude/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .agents/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .cursor/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .gemini/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .github/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add github/gh-aw-firewall --skill add-llm-provider -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "add-llm-provider" agent skill from https://github.com/github/gh-aw-firewall/tree/main/.github/skills/add-llm-provider into .opencode/skills/add-llm-provider/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-llm-provider", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
add-llm-provider
GitHub stars
148
Token cost
~1.4k tokens
SKILL.md length
653 words
Files
1
Skills in repo
6
Repo updated
First seen
Licence
MIT
At a glance
Decide and implement how to support a new LLM provider or agent engine in AWF - either as a proxied provider (api-proxy adapter) or a direct-API engine (domain allowlist only), e.g.
Works in 7 steps: Create… → Register it in… → Add its port to… → …
SKILL.md covers Step 1 — Determine how the…, Path A — Proxied provider…, Path B — Direct-API engine (no… and Checklist
Calls docker and npm
What it does
Add LLM Provider is an agent skill from github/gh-aw-firewall, published by the product's own GitHub organization. Decide and implement how to support a new LLM provider or agent engine in AWF - either as a proxied provider (api-proxy adapter) or a direct-API engine (domain allowlist only), e.g. Cursor, Aider, or any tool that calls its own endpoint.
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It works with GitHub. The repository describes itself as: GitHub Agentic Workflows Firewall. The licence is MIT.
Example prompts
“/add-llm-provider”
Requirements
Docker
Workflow steps
7 steps, taken from the first numbered list in SKILL.md.
1Create containers/api-proxy/providers/.js implementing the ProviderAdapter interface.
2Register it in containers/api-proxy/providers/index.js.
3Add its port to src/config/sandbox-network-policy.json, then extend the closed NetworkPolicy shape and validation in…
4Add each credential to the API-proxy environment pipeline in src/services/api-proxy-env-config.ts and explicitly exclude it from agent…
5Add the upstream domain(s) to the allowlist wherever the caller configures --allow-domains (AWF itself does not hardcode per-provider…
6Document auth details in docs/auth-matrix.md if it's a net-new auth pattern.
7Write adapter unit tests (providers/.test.js) and run cd containers/api-proxy && npm test -- providers/.test.js.
What it can do on your machine
Read from SKILL.md and the folder at commit e51e3ce. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
docker
npm
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use docker and npm, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Add LLM Provider loads about 1.4k tokens when it runs. Until then it costs about 64 tokens; SKILL.md has 653 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~64
When it runs· the whole SKILL.md, loaded when a task matches
~1.4k
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Download SKILL.mdSave it as .claude/skills/add-llm-provider/SKILL.md (or your agent's skills folder).
name
add-llm-provider
description
Decide and implement how to support a new LLM provider or agent engine in AWF - either as a proxied provider (api-proxy adapter) or a direct-API engine (domain allowlist only), e.g. Cursor, Aider, or any tool that calls its own endpoint.
Add LLM Provider
Use this skill whenever a new agent engine or LLM provider needs to work behind AWF and it is unclear how to "set it up." AWF supports two fundamentally different integration paths — picking the wrong one is the most common source of confusion (e.g. "why is token-usage.jsonl empty for this engine?").
Step 1 — Determine how the engine talks to its LLM backend
Ask: does the engine call the provider's API directly (its own base URL, its own auth), or can it be configured to route through AWF's API-proxy sidecar? The sidecar is always enabled; --enable-api-proxy is deprecated and ignored. Its provider ports are http://172.30.0.30:10000 (OpenAI), :10001 (Anthropic), :10002 (Copilot), :10003 (Gemini), and :10004 (Vertex AI).
If the engine has built-in support for pointing at AWF's api-proxy ports (env vars like OPENAI_BASE_URL, ANTHROPIC_BASE_URL rewritten to the sidecar) → Path A: Proxied provider.
If the engine always calls its own hardcoded/native endpoint regardless of proxy env vars (e.g. Cursor CLI calling api2.cursor.sh / api3.cursor.sh directly), or manages its own credentials outside AWF → Path B: Direct-API engine.
When unsure, run the engine once with --keep-containers and inspect docker exec awf-squid cat /var/log/squid/access.log (see docs/squid_log_filtering.md) to see which hosts it actually contacts and whether traffic reaches the api-proxy sidecar or goes straight out through Squid.
Path A — Proxied provider (credentials injected by api-proxy)
Choose this when you want AWF to hold the API key/OIDC token and inject it, keeping secrets out of the agent container.
Create containers/api-proxy/providers/<name>.js implementing the ProviderAdapter interface.
Register it in containers/api-proxy/providers/index.js.
Add its port to src/config/sandbox-network-policy.json, then extend the closed NetworkPolicy shape and validation in src/config/network-policy.ts and the compatibility mapping in src/types/ports.ts. Add it to the Dockerfile EXPOSE list; the whole providers/ directory is already copied. src/host-iptables-rules.ts consumes Object.values(API_PROXY_PORTS), so it normally needs no change.
Add each credential to the API-proxy environment pipeline in src/services/api-proxy-env-config.ts and explicitly exclude it from agent passthrough in src/services/agent-environment/excluded-vars.ts; never forward a provider credential generically or from src/docker-manager.ts.
Add the upstream domain(s) to the allowlist wherever the caller configures --allow-domains (AWF itself does not hardcode per-provider domains).
Document auth details in docs/auth-matrix.md if it's a net-new auth pattern.
Write adapter unit tests (providers/<name>.test.js) and run cd containers/api-proxy && npm test -- providers/<name>.test.js.
Show full SKILL.md (278 more words)Show less
Path B — Direct-API engine (no proxy adapter needed)
Choose this when the engine calls its own API directly and either manages its own credentials, or credential injection isn't feasible/needed through AWF.
No API-proxy adapter is required. The API proxy remains enabled, but do not try to force an engine through it when it cannot use the sidecar.
Just allowlist the domain(s) the engine needs, either via the CLI flag or config file:bash
If the engine requires its API key in the agent environment, it cannot receive API-proxy credential isolation. Pass only a minimally scoped credential by an intentional mechanism appropriate to the caller, and do not use sensitiveAllowedDomains for the key: that setting only redacts secret-derived endpoint hostnames in logs and audit artifacts.
Telemetry caveat: because traffic never passes through the API-proxy sidecar, AWF's token-usage.jsonl / token-tracking metrics will stay empty for this engine. Any downstream check that assumes all engines produce proxy telemetry (e.g. a "token usage present" CI gate) must exclude direct-API engines instead of trying to make them populate it.
Confirm the domains are correct by testing with --keep-containers and checking Squid's access log for TCP_DENIED entries, then iterating on the allowlist (see docs/quickstart.md "Test Domain Blocking" section).
Checklist
Identified whether the engine is proxied (Path A) or direct-API (Path B)
Path A: adapter created and registered; network policy, credential isolation, image port, and tests updated
Path B: domain(s) allowlisted, telemetry-dependent checks updated to exclude this engine
Verified with --keep-containers + Squid access log that the engine's real traffic is allowed and nothing extraneous is
Add LLM Provider next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Investigates a session where Superpowers went wrong, reads the transcripts on disk and produces an evidence-cited report, optionally prepared as a bug report for the maintainers.
Researches a GitHub repository over four rounds using the GitHub API and web search, then writes a structured markdown report with timeline, metrics and Mermaid diagrams.
Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.
Bumps the pinned v8 and rusty_v8 versions in Codex, validates the release-candidate path with the v8-canary check, and traces failures to upstream build changes.
Decide and implement how to support a new LLM provider or agent engine in AWF - either as a proxied provider (api-proxy adapter) or a direct-API engine (domain allowlist only), e.g. Add LLM Provider is an agent skill from github/gh-aw-firewall, published by the product's own GitHub organization.g.
How do I install Add LLM Provider in Claude Code?
Run `npx skills add github/gh-aw-firewall --skill add-llm-provider -a claude-code`. Or copy the skill folder (.github/skills/add-llm-provider in github/gh-aw-firewall) into .claude/skills/add-llm-provider in your project. Claude Code loads it when a task matches its description.
How do I install Add LLM Provider in Codex?
Run `npx skills add github/gh-aw-firewall --skill add-llm-provider -a codex`. Or copy the skill folder (.github/skills/add-llm-provider in github/gh-aw-firewall) into .agents/skills/add-llm-provider in your project. Codex loads it when a task matches its description.
Can I use Add LLM Provider in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add github/gh-aw-firewall --skill add-llm-provider -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/add-llm-provider, .gemini/skills/add-llm-provider, .github/skills/add-llm-provider and .opencode/skills/add-llm-provider in your project.
What does Add LLM Provider need to run?
Going by SKILL.md and its folder, Add LLM Provider needs the command-line tools its instructions call (docker and npm). Our summary lists: Docker.
Does Add LLM Provider access the network?
SKILL.md contains no URLs. Its commands use docker and npm, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Add LLM Provider safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Add LLM Provider use?
Add LLM Provider is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Add LLM Provider use?
About 1.4k tokens (SKILL.md is roughly 5.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Add LLM Provider?
Skills that share tags, products or a category with Add LLM Provider: PR Babysitter (openinterpreter/openinterpreter, 69k stars), Diagnosing Superpowers Sessions (obra/superpowers, 296k stars), GitHub Deep Research (bytedance/deer-flow, 83k stars) and Greploop (onyx-dot-app/onyx, 32k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Add LLM Provider?
github (a GitHub organization, an official publisher) maintains it in github/gh-aw-firewall, which has 148 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on October 8, 2026.
Source: github/gh-aw-firewall on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.
