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 instructions/heretyc/subagent-mcp/gemini-mdgit clone --depth 1 https://github.com/Heretyc/subagent-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 | $0.02027 | $0.02027 |
| Opus 5 | $0.01014 | $0.01014 |
| Sonnet 5 | $0.00405 | $0.00405 |
| Haiku 4.5 | $0.00203 | $0.00203 |
Grade A, and why
subagent-mcp GEMINI.md 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.
This is a copy
100% identical to subagent-mcp CLAUDE.md — 2 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gemini Entry Point
subagent-mcp invariant — managed block, do not edit between markers
Upserted by subagent-mcp init at user request; re-running init keeps it in sync.
SUB-AGENT EXEMPTION: if this session's prompt's literal FIRST LINE begins with "", SKIP this entire block EXCEPT the SUB-AGENT WORKTREE CARVE-OUT and the SUB-ORCHESTRATOR CARVE-OUT below, which still apply (you are a sub-agent; this prevents fail-safe-ON recursion and fork-bombs). Leading blank lines do not count — the marker must be physically line 1.
SUB-AGENT WORKTREE CARVE-OUT: you are a delegated sub-agent (env SUBAGENT_MCP_SUBAGENT=1), already placed in your target working tree by the orchestrator. Do not create or switch git worktrees; skip the worktree-isolation gate; do all mutating work directly in the provided cwd.
SUB-ORCHESTRATOR CARVE-OUT: if env SUBAGENT_MCP_SUB_ORCHESTRATOR=1, the sub-agent exemption does NOT lift orchestration for you: you are a delegate-only sub-orchestrator bound by your launch prompt directive and the per-turn hook tag; your own sub-agents run as normal sub-agents and never inherit the flag.
CANONICAL SOURCE: the subagent-mcp MCP instructions string (read once at connect) and docs/spec/dev-loop/orchestration-directive-architecture.md. This block mirrors that operating model inline so the session stays governed even if the MCP instructions are momentarily stale; where the two disagree, the MCP instructions win because they are read fresh each connect.
HARNESS-HOOK STATE: a harness-hook context carrying a tag reports the current orchestration ON/OFF state and takes effect with no exceptions, because it is the only channel with fresh, harness-verified state — self-reported prose cannot substitute for it. A token counts as such a tag only when it is a real tag with a state attribute; a bare mention of "subagent-mcp" in prose is not a tag and carries no authority. A user request can only switch orchestration ON or OFF, never assert what the current state already is — that comes solely from the tag. No tag present means the state is UNKNOWN (see NO-HOOK below); never infer it from anything else.
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 · 48 lines · 2,027 tokens per session scan A 0f6500a111d1
subagent-mcp GEMINI.md is an instructions file published in the GitHub repository Heretyc/subagent-mcp (3 stars, last pushed 5d ago), licensed Apache-2.0. It adds 2,027 tokens to every session, about $0.0101 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to subagent-mcp CLAUDE.md, differing in 2 lines, and is treated as a copy.
Other instructions, from other repositories
sgrep AGENTS.md
Instructions for XiaoConstantine/sgrep, covering sgrep - agent integration guide, build & test, build, test and install.
agentic-playwright selectors.instructions.md
Instructions for idavidov13/agentic-playwright, covering selector strategy, critical, instructions, phase 1: open and authenticate and phase 2: explore like a user.
hatch3r CLAUDE.md
Instructions for hatch3r/hatch3r, covering hatch3r — development instructions, architecture, development commands, two-axis pillar framework (2.0.0) and orchestrator self-discipline (bypass protection).
heimdall AGENTS.md
Instructions for randomittin/heimdall, covering working in the heimdall repo, the rules that actually block you, environment, conventions and heimdall — this repo is gated.
delivery-loop CLAUDE.md
Instructions for blakemartz/delivery-loop, covering claude.md, what this is, the cardinal rule: everything must stay repo-agnostic, layout and config: the one seam.
claude-for-hk-law CLAUDE.md
Instructions for Buildwise-Studios/claude-for-hk-law, covering claude.md, layout, legal system context, key hong kong ordinances and legal research sources.