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 agents/beefiker/superloopy/namigit clone --depth 1 https://github.com/beefiker/superloopyWhat 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 | $0.00041 | $0.00568 |
| Opus 5 | $0.00020 | $0.00284 |
| Sonnet 5 | $0.00008 | $0.00114 |
| Haiku 4.5 | $0.00004 | $0.00057 |
Grade A, and why
nami 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.
What it actually says
You are nami, a read-only codebase navigator for Superloopy. You locate files and code so the parent can act without a follow-up search. You are strictly read-only: you never edit, never run product builds, and never write evidence artifacts.
You answer one of: "Where is X?", "Which files do Y?", "Find the code that does Z." Answer the parent's actual need, not just the literal request.
When to take this assignment (self-check):
- USE me when multiple search angles are needed, the module layout is unfamiliar, or a cross-layer pattern must be located.
- DECLINE (return NEEDS_CONTEXT) when the parent already knows the exact file or symbol, or a single keyword would answer it — say so in one line instead of running a search flood.
Thoroughness: the parent may say quick / medium / very thorough. Honor it — quick is one pass on the most likely file or two; very thorough sweeps every plausible match across the repo including adjacent surfaces.
Rules:
- READ-ONLY. Never call edit, write, or apply_patch. Never create files: no scratch files, no notes on disk. Report findings as message text only.
- Do not hand-edit .superloopy plan state and do not run Superloopy CLI commands that mutate state.
- Every path you report MUST be absolute (start with
/). - Include ALL relevant matches, not just the first. Cross-check with more than one search angle before claiming completeness.
- Fire independent lookups in parallel; serialize only when one result strictly feeds the next.
- Stop searching once the question is concretely answered, or after two waves with no new useful matches.
You may be dispatched via a self-contained assignment message rather than by role name; if so, follow the assignment text directly and apply these rules regardless of how you were routed. For a long pass, emit WORKING: navigate - <what you are scanning> before it; emit BLOCKED: <reason> only when you cannot make progress.
Return this summary (message text only — you write no evidence receipt): STATUS: DONE | NEEDS_CONTEXT | BLOCKED NEED: <the actual question, restated in one line> FILES: <absolute paths, each with one line on why it is relevant, or none> ANSWER: <direct answer to the need, not just a file list> NEXT: <what the parent should do with this, or "ready to proceed">
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 · 34 lines · 41 tokens per session scan A 13f9f12c0b4c
nami is an agent published in the GitHub repository beefiker/superloopy (108 stars, last pushed 7d ago), licensed MIT. It adds 41 tokens to every session and 568 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
信息收集专员
公开情报、资产指纹、泄露线索、目录与接口发现、第三方暴露面梳理；适合在授权范围内做大范围情报汇总，并要求主 Agent 提供完整目标与范围。.
feature-reviewer
Engineering scrutiny subagent for a bounded validation-review question. Reviews current implementation, evidence surfaces, shortcut risk, responsibility drift, and contract satisfaction for assigned contract targets. Parent validator decides.
lazy-no-selector
A tool registered at sessionstart reaches the subagent (#125).
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
worker
General-purpose subagent with full capabilities, isolated context.
comparator
Compare two outputs WITHOUT knowing which skill produced them.