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/sandyup/comfyui-mcp/convertgit clone --depth 1 https://github.com/sandyup/comfyui-mcpWhat 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 | $0.00011 | $0.00968 |
| Opus 5 | $0.00005 | $0.00484 |
| Sonnet 5 | $0.00002 | $0.00194 |
| Haiku 4.5 | $0.00001 | $0.00097 |
Grade A, and why
convert 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/comfy-convert — Convert Workflow Format
The user wants to convert a ComfyUI workflow between the web UI format and the API format.
Instructions
-
Read the workflow file. The argument is: $ARGUMENTS
- If a file path is provided, read it with the Read tool
- If no argument is provided, ask the user for a file path or to paste the JSON
-
Detect the format. Examine the JSON structure:
- UI format: has top-level
"nodes"array and"links"array (this is what "Save" in the web UI produces) - API format: has string keys (node IDs like
"1","2") where each value has"class_type"and"inputs"(this is what "Save (API Format)" produces)
Tell the user which format was detected and which format it will be converted to.
- UI format: has top-level
-
If converting UI format to API format:
a. Iterate over each node in the
"nodes"array. Each node has:id(integer) — becomes the string key in API formattype— becomesclass_typewidgets_values— ordered list of widget valuesinputs— array of input slots withname,type, and optionallinkoutputs— array of output slots
b. For each node, call
get_node_infowith the node'stypeto get the input schema. This tells you:- The order and names of widget inputs (to map
widgets_valuescorrectly) - Which inputs are required vs optional
- Expected types for each input
c. Build the API node's
inputsobject:- Map
widgets_valuesto named inputs using the order fromget_node_info - For linked inputs (where
linkis not null), find the connection in the"links"array. Each link is[link_id, source_node_id, source_slot, target_node_id, target_slot, type]. Set the input value to["{source_node_id}", source_slot]
d. Assemble the API workflow:
{ "{node_id}": { "class_type": type, "inputs": {...} }, ... } -
If converting API format to UI format:
Warn the user that API-to-UI conversion is lossy — layout positions, colors, groups, and notes will not be preserved. The converted workflow will load in ComfyUI but nodes will need to be manually arranged.
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 · 79 lines · 11 tokens per session scan A 4b27fc160bf2
convert is a command published in the GitHub repository sandyup/comfyui-mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 11 tokens to every session and 968 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
brand-generate
Generate an on-brand document from a saved Brand Profile.
stt
Transcribe a local audio file or remote audio URL into text.
audition-voices
Generate voice audition samples for a character using Venice TTS.
search
Search videos by text, image, or entity using TwelveLabs.
love
Love the current track — strong positive signal for this vibe.
speckit.tasks
Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.