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.
git clone --depth 1 https://github.com/tboome33/openrouter-fusion-mcpWrote 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/commands/tboome33/openrouter-fusion-mcp/fusion)<a href="https://agentmods.dev/commands/tboome33/openrouter-fusion-mcp/fusion"><img src="https://agentmods.dev/badge/commands/tboome33/openrouter-fusion-mcp/fusion/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/commands/tboome33/openrouter-fusion-mcp/fusion"><img src="https://agentmods.dev/badge/commands/tboome33/openrouter-fusion-mcp/fusion.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 | $0.00026 | $0.00619 |
| Opus 5 | $0.00013 | $0.00309 |
| Sonnet 5 | $0.00005 | $0.00124 |
| Haiku 4.5 | $0.00003 | $0.00062 |
Grade A, and why
fusion scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Drive OpenRouter Fusion interactively. Follow these steps in order, stopping to wait for the user's answer after each question (don't chain ahead).
1. List the configs. Call fusion_list. Present EVERY preset returned as a full TABLE (one
row per preset; columns: #, preset, est. cost ~$low–$high (from cost_estimate.low/.high),
panel, judge/orchestrator, reasoning_effort). ALWAYS render this full table at the moment
you ask the user to choose — even if you showed it earlier, and even when you recommend one: never
reduce it to just the recommendation. The cost is an estimated RANGE per run (floor = little web,
ceiling = full max_tool_calls web budget); real cost also scales with prompt size (each preset's
cost_estimate.usd_per_prompt_token) and reasoning effort. State it's indicative, not a quote.
fusion_startis gated by a permission confirmation in Claude Code (the user approves each paid launch). You don't manage it — the harness prompts.
2. Recommend + ask for the preset. Mark (⭐) the preset best suited to the request (with its
cost_tier), then ask "Which preset (number or name)?" and WAIT. Never pick for them. (Skip this
ONLY if the user named the preset themselves in the request — you proposing one does not count.)
3. Ask for the reasoning effort — xhigh · high · medium · low · minimal · none, recalling the
chosen preset's default. Then WAIT.
4. Ask for the temperature — a number 0–2, or "model default". Then WAIT.
5. Ask for the question if it wasn't already provided.
6. Run. Call fusion_start with preset, prompt (the full question), and
reasoning_effort / temperature only if the user chose an explicit value (otherwise omit them).
Get the job_id.
7. Poll. Call fusion_result with that job_id; while {status:"running"}, call again with
the same job_id until the final answer (~45 s long-poll per call).
8. Return the synthesized answer verbatim (cost footer included), without reformulating.
On error (unknown preset, expired job_id, 401…), explain briefly and don't loop.
Request (may be empty — then ask for it at step 5): $ARGUMENTS
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.
- 11d ago First seen · 44 lines · 26 tokens per session scan A af59aa50c7c8
fusion is a command published in the GitHub repository tboome33/openrouter-fusion-mcp (0 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 619 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
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.