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/willackerly/rebar/subagent-guidelinesgit clone --depth 1 https://github.com/willackerly/rebarWhat 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 | $0.00000 | $0.02137 |
| Opus 5 | $0.00000 | $0.01069 |
| Sonnet 5 | $0.00000 | $0.00427 |
| Haiku 4.5 | $0.00000 | $0.00214 |
Grade A, and why
subagent-guidelines 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 — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Subagent Guidelines
Shared behavioral contract for all subagent invocations. The orchestrating agent should instruct every subagent to read this file before starting work.
Context Loading Order
When you start, read files in this order:
- This file (
agents/subagent-guidelines.md) — you're here - Your assigned template from
agents/subagent-prompts/<template>.md - QUICKCONTEXT.md — project orientation (if your task touches project code)
- Any additional files specified in your parameters
Do not skip steps. Do not start working before reading your template.
The 10 Rules (non-negotiable)
These rules are the difference between 100% work survival and total loss. They are numbered for reference — cite "Rule 3" in findings, not prose.
Rule 1: Worktree Isolation for Code Changes
If you are writing ANY code, you MUST use worktree isolation. No exceptions. Read-only research tasks (exploration, analysis, audits) may run in the main working tree.
rebar agent start --role developer "implement feature X"
Rule 2: Commit After Every Logical Chunk
Don't accumulate changes. If you get logged out, only uncommitted work is
lost. One fix = one commit. Use rebar commit (not git commit) — it
runs pre-commit checks and updates integrity. There is no --no-verify.
rebar commit -m "agents/<template-name>: <brief description>"
Rule 3: Strict File Ownership
Your prompt includes an allowlist of files you may create or modify. Everything else is read-only. If your prompt doesn't specify, you may modify files directly related to your assigned task. You may NOT modify:
- Shared interface definitions (Store, Scanner, etc.)
- Shared type files, routers, app entry points
- Migration files, lock files, auto-generated files
If you need a change to a restricted file, document it as a finding.
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 · 248 lines · 0 tokens per session scan A 4967b6295baa
subagent-guidelines is an agent published in the GitHub repository willackerly/rebar (5 stars, last pushed 20d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,137 tokens. 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 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.
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…
worker
General-purpose subagent with full capabilities, isolated context.