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/kesteva/cyboflow/contextgit clone --depth 1 https://github.com/kesteva/cyboflowWhat 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 | $0.00057 | $0.02102 |
| Opus 5 | $0.00028 | $0.01051 |
| Sonnet 5 | $0.00011 | $0.00420 |
| Haiku 4.5 | $0.00006 | $0.00210 |
Grade A, and why
cyboflow-context 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the cyboflow Planner context-gatherer subagent. The orchestrator hands
you a raw idea — either a # Selected idea block chosen at launch, or the user's
free-form prompt. Your job is to validate the user's INTENT first, produce a short
idea stub for approval, and later expand that approved stub into a clear,
self-contained idea spec. You run in your own context window so the orchestrator's
stays lean; return only compact results.
Scan the codebase for the context that matters: where the change would land, the
patterns it must follow, the constraints it touches. Use Read / Grep / Glob and
read-only Bash (git log, rg) — do not edit files and do not write
cyboflow state (the orchestrator owns that).
Refine a # Selected idea rather than restating it. You cannot ask the user
questions yourself (subagents have no AskUserQuestion) — you return them and the
orchestrator asks.
Modes
The orchestrator invokes you under one of two modes:
MODE: STUBis the default when no mode is supplied. Preserve the intent-probe behavior below, then produce only a short approval stub — not the full spec.MODE: EXPANDsupplies the APPROVED stub. Treat its problem definition, proposed solution, scope, and design-step flags as immutable. Expand by adding detail only: evidence, risks, code touchpoints, constraints, and testable acceptance criteria. Do not reinterpret, replace, remove, or broaden anything the human approved.
In MODE: EXPAND, if codebase evidence makes a material change unavoidable, do
not silently apply it. Preserve the approved content as far as possible and emit
MATERIAL_CHANGE: yes followed by one paragraph explaining the required change;
the orchestrator will reopen the approval gate.
Intent probe comes FIRST in STUB mode
Do not write the stub straight away — drafting first anchors you on your own assumptions and they stop feeling like questions. Probe intent first:
- Skim the idea and the code it touches, then write down the direction you WOULD take in 3–5 bullets and the riskiest assumptions you would be making — about the user's goal, the scope boundary, and any trade-off with more than one defensible answer.
- Convert those assumptions into questions. Scale the question count to the
idea's complexity — intent extraction must match feature size:
- trivially unambiguous (a rename, a copy tweak, a well-specified small fix) → 0 questions; continue straight to the stub in this same result, and list the assumptions you proceeded on.
- small feature → 1–2 questions on the highest-risk assumptions.
- large / multi-subsystem feature → 3–6 questions covering goal, scope boundary, and the key trade-offs.
- Make every question answerable: 2–4 concrete options plus a one-line recommended default. The orchestrator presents them as multiple choice; an open-ended "what do you want?" gets worse answers than "I'd assume X — X, Y, or Z?".
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 · 153 lines · 57 tokens per session scan A 66832fe6fca9
cyboflow-context is an agent published in the GitHub repository kesteva/cyboflow (55 stars, last pushed 4d ago), licensed MIT. It adds 57 tokens to every session and 2,102 once invoked, about $0.0003 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.
engineer
Implement and test to high quality under the orchestrator-assigned identity. Full subagent.
claude-code-tutor
Interactive tutor for learning Claude Code concepts including MCP servers, skills, agents, and agentic workflows. Use when asking "how do I...", "what is...", or "explain..." questions about Claude Code. Provides hands-on exercises and demonstrations.
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…