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 agentmods add commands/nexenio/bifrost-plugin/bifrost-setupgit clone --depth 1 https://github.com/neXenio/bifrost-pluginWhat 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 | $0.00021 | $0.01010 |
| Opus 5 | $0.00010 | $0.00505 |
| Sonnet 5 | $0.00004 | $0.00202 |
| Haiku 4.5 | $0.00002 | $0.00101 |
Grade A, and why
bifrost-setup 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 2d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/bifrost-setup
Set up a Bifrost MCP gateway and confirm memory and skill discovery are live. This command is the ONLY place onboarding runs. The plugin never launches setup on its own.
Primary path: install the plugin
CLI: /plugin marketplace add neXenio/bifrost-plugin then /plugin install bifrost-plugin. On Desktop and claude.ai, see the per-surface install steps
in the bifrost-onboard skill.
Installing prompts for the plugin's three config values:
- Gateway URL (
gateway_url): defaults to the shared gateway. - Virtual key (
virtual_key, optional): paste thevk_...key your gateway operator issued you. Leave it blank to sign in with your company account through OAuth instead. - OAuth client ID (
oauth_client_id, optional): only needed if your identity provider does not let Claude register itself. Ask your gateway operator for it.
The bundled .mcp.json reads these back as ${user_config.gateway_url} and so
on, so no separate registration step is needed. Change any value later with
/plugin configure, no reinstall required.
What this command does
- Checks whether the plugin is installed and enabled (
/plugin). - If not, walks you through install and the config prompts above.
- Guides you to verify the connection.
Env vars: override for the hook layer only
BIFROST_URL and BIFROST_VK, if both set in your shell, override the
plugin's configured gateway URL and virtual key for the hook layer only
(session-start.cjs, prompt-submit.cjs, session-reflect.cjs,
usage.cjs). Use this to point hooks at a different gateway than the one
configured in the plugin, or when running the CLI without a plugin install at
all. The MCP connection itself always uses the plugin's config when the
plugin is installed; these env vars do not change it.
If you are not using the plugin at all, you can still register the server directly with the Claude Code CLI:
export BIFROST_URL=https://<your-gateway-host>/mcp
node "${CLAUDE_PLUGIN_ROOT}/bin/install.js" --key vk_<your-key>
# Or without a key (VK must already be in env):
node "${CLAUDE_PLUGIN_ROOT}/bin/install.js"
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.
- 2d ago First seen · 97 lines · 21 tokens per session scan A 55f41283ae19
bifrost-setup is a command published in the GitHub repository neXenio/bifrost-plugin (2 stars, last pushed 14d ago), licensed MIT. It adds 21 tokens to every session and 1,010 once invoked, about $0.0001 per session on Opus 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 2026-08-31.
Other commands, from other repositories
gen-agent
You are a specialized prompt engineer tasked with generating high-quality, structured prompts for AI agents based on user descriptions. Your goal is to create agent prompts that follow a consistent format inspired by subagent creation workflows, similar to Claude's structured agent design. When you receive an input in…
continue
Continue last CCS delegation session [AUTO ENHANCE].
models
Alias for /router-models — list the models the Weave Router may route to, and turn them on or off.
router-models
List the models the Weave Router may route to, and turn them on or off.
router-session
Print the session id that correlates this session in router telemetry and logs.
router-off
Route Claude Code directly to Anthropic again (turn the Weave Router off).