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/copilot-instructionsgit 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 | $0.00726 | $0.00726 |
| Opus 5 | $0.00363 | $0.00363 |
| Sonnet 5 | $0.00145 | $0.00145 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
subagent-mcp copilot-instructions.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.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Repository Instructions
subagent-mcp turns a host CLI into an orchestrator for local sub-agent
sessions. It exposes MCP tools such as launch_agent, poll_agent, wait,
get_status, orchestration-mode, swarm, and handoff tools. Provider
credentials and user config live outside the repo.
The sub-orchestrator: true flag on launch_agent is for the swarm dispatch
stage only (main orchestrator, depth 0); it launches a delegate-only child
orchestrator for one disjoint plan section and is never set elsewhere.
Fallback Rule
Use this file when the host reads instruction files but does not load subagent-mcp skills, hooks, or slash commands. This applies to instruction-tier fallback use in Gemini CLI, Cursor, Windsurf, Kiro, Copilot, and similar agents.
Orchestration
- If a
<subagent-mcp state="...">hook tag is present, treat it as the authoritative orchestration state. - If no hook tag is present, say:
subagent-mcp: no hook injection detected: orchestration state unknown; defaulting to ON. - When orchestration is ON, use only
launch_agentfor sub-agent work. Do not use native agent/task tools or shell-spawned agents. - Every sub-agent prompt must begin with
<this is a request from a parent process>. - Sub-agents return JSON with
status,summary,source_locators,risks, andwrites_requested.
Read Ladder
- Use
poll_agenttail output first. - If the tail is insufficient, launch one summarizer sub-agent and trust its summary if it is 100 lines or fewer.
- For larger handoffs, have sub-agents write scratch files and pass file paths between them. The orchestrator does not read large files inline.
- Use
waitto learn completion. A quiet or stalled agent is still alive.
Status And Repair
/smcp:help: loadskills/smcp-help/SKILL.mdfor install, config, and maintenance guidance./smcp:status: loadskills/smcp-status/SKILL.mdand callget_status./smcp:doctor: loadskills/smcp-doctor/SKILL.md; runsubagent-mcp doctorread-only first and ask before any repair./smcp:config: loadskills/smcp-config/SKILL.mdand callconfigureto list, read, or update configuration by canonical key.- Config commands:
subagent-mcp config init,subagent-mcp config validate,subagent-mcp doctor,subagent-mcp upgrade,subagent-mcp rollback.
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 · 61 lines · 726 tokens per session scan A 599743e20109
subagent-mcp copilot-instructions.md is an instructions file published in the GitHub repository Heretyc/subagent-mcp (3 stars, last pushed 4d ago), licensed Apache-2.0. It adds 726 tokens to every session, about $0.0036 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 instructions, from other repositories
neo4j-skills AGENTS.md
Instructions for neo4j-contrib/neo4j-skills, covering writing skills for this repository, language style — write for agents, not humans, six design principles, skill.md spec (agentskills.io) and frontmatter.
sgrep AGENTS.md
Instructions for XiaoConstantine/sgrep, covering sgrep - agent integration guide, build & test, build, test and install.
harness-anchor CLAUDE.md
Instructions for Redtropig/harness-anchor, covering harness-anchor — contributor & ai collaborator guidelines, if you are an ai agent editing this plugin, design invariants (do not break), shell hazards no tool in this repo catches and authoring a new skill (when explicitly asked).
awesome-gemini-cli-subagents GEMINI.md
Instructions for ankitmundada/awesome-gemini-cli-subagents, covering gemini.md, project overview, repository structure, subagent file format and communication protocol.
linkedin-maxxing CLAUDE.md
Claude Code instructions for warpirate/linkedin-maxxing, covering claude.md, what this repo is, layout, skill anatomy and how writing skills produce non-ai output.
redthread AGENTS.md
AGENTS.md instructions for matheusht/redthread, covering redthread agent architecture, 1. core operating guidelines, 2. the orchestration workflow (principal vs subagents), the principal agent must and subagent usage.