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/quantumwars/project-graphx/setupgit clone --depth 1 https://github.com/QuantumWars/project-graphxWhat 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 | $0.00025 | $0.00950 |
| Opus 5 | $0.00013 | $0.00475 |
| Sonnet 5 | $0.00005 | $0.00190 |
| Haiku 4.5 | $0.00003 | $0.00095 |
Grade A, and why
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 yesterday.
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.
Create .claude/graph/config.json for the current project. Do not invent its contents — find them.
1. Find the sources
A source is a directory that holds agents or skills. Look for these, in the project and anywhere it vendors other repos:
.claude/agents/and.claude/skills/— the project's own- any
agents/orskills/directory belonging to a plugin or a vendored catalogue in this repo
Use Glob for **/.claude/agents, **/.claude/skills, **/skills/*/SKILL.md, **/agents/*.md, and prune node_modules. Report what you found with counts before writing anything.
Each source becomes one entry:
{ "repo": "<label shown in the graph>", "root": "<path relative to project root>", "kind": "agent" | "skill" }
kind is agent for a directory of *.md files, skill for a directory of <name>/SKILL.md directories. A repo with both needs two entries.
2. Choose the scan roots
scanRoots are the trees searched for projects that have installed these skills — that is what fills in "who actually uses this". Each is a directory holding many repos, e.g. ~/code. Ask the user which to use rather than guessing; [] is a legitimate answer meaning "catalogue only, do not scan".
scanExclude drops paths whose text contains any listed substring. Always keep /node_modules/ and /imported-repos/ — the second keeps the catalogue's own add_repo clones, which carry a .claude/ of their own, from being counted as your projects. Add the source trees themselves if they are study material rather than projects — otherwise the catalogue counts itself as a user of everything in it.
3. Write it
{
"sources": [ ... ],
"scanRoots": ["~/code"],
"scanExclude": ["/node_modules/", "/imported-repos/"]
}
Write to .claude/graph/config.json, creating the directory. If a config already exists, show the user the difference and confirm before overwriting.
4. Show it, then offer to build
Show the config you wrote, then ask whether to build now and run it if they agree. Do not
simply end with an instruction to go and run /skill-graph:build — a config with no graph behind
it does nothing, and the user has no way to judge whether the sources were right until they see
what comes out of them.
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.
- yesterday First seen · 79 lines · 25 tokens per session scan A 3dee09353c20
setup is a command published in the GitHub repository QuantumWars/project-graphx (1 stars, last pushed 13d ago), licensed MIT. It adds 25 tokens to every session and 950 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.