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/userb1ank/interactive-process-mcp/claude-mdgit clone --depth 1 https://github.com/UserB1ank/interactive-process-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 | $0.00597 | $0.00597 |
| Opus 5 | $0.00298 | $0.00298 |
| Sonnet 5 | $0.00119 | $0.00119 |
| Haiku 4.5 | $0.00060 | $0.00060 |
Grade A, and why
interactive-process-mcp CLAUDE.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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Security & Privacy
禁止在项目中出现以下敏感信息和敏感文件：
- 本机 hostname、本地用户名称
- 真实密钥、API keys、tokens、passwords
- 邮箱地址、手机号等个人信息
- .env 文件、私钥文件（PEM、SSH key 等）
- 禁止出现
Co-Authored-By: Claude Opus 4.7 <[email protected]>相关字样
Agent skills
Issue tracker
Issues are tracked in GitHub Issues via gh CLI. See docs/agents/issue-tracker.md.
Triage labels
Default vocabulary: needs-triage, needs-info, ready-for-agent, ready-for-human, wontfix. See docs/agents/triage-labels.md.
Domain docs
Single-context layout: CONTEXT.md + docs/adr/ at repo root. See docs/agents/domain.md.
Multi-session parallel work
Agents using interactive-process MCP tools must follow these rules for non-blocking, multi-session operation.
Rules
- One task = one session. Start a new session per independent task via
start_process. Use thenameparam for tracking. - Never block on reads. Always use
read_outputwithtimeout≤ 3. A long timeout blocks the entire agent — other sessions go unserviced. - Prefer
send_and_read. For send-then-read patterns, usesend_and_readinstead of separatesend_input+read_outputcalls. Same timeout rule applies. - Poll in rotation. When managing N sessions, loop through all of them:
read_output(timeout=1)each, act on whichever has output, repeat. - Clean up.
terminate_processthendelete_sessionwhen done. Never leave zombie sessions.
Multi-agent shared session
When multiple agents need to observe the same process:
- Agent A:
start_process(...)→ session_id, default reader_id=0 - Agent B:
register_reader(session_id=...)→ gets its own reader_id - Each agent calls
read_output(session_id=..., reader_id=<theirs>)— independent cursors, no output stealing - Agent B leaves:
unregister_reader(session_id=..., reader_id=...)
Anti-patterns
- ❌
read_output(timeout=30)— blocks 30s, other sessions starve - ❌ Waiting for session A to finish before starting session B — start both, poll both
- ❌ Multiple agents using the same reader_id — output gets consumed, others miss it
- ❌ Forgetting
delete_sessionafterterminate_process— stale metadata accumulates
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 · 52 lines · 597 tokens per session scan A 0170c46c5be3
interactive-process-mcp CLAUDE.md is an instructions file published in the GitHub repository UserB1ank/interactive-process-mcp (8 stars, last pushed 3mo ago), licensed MIT. It adds 597 tokens to every session, about $0.0030 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
termcp CLAUDE.md
Instructions for open-mcp-ai/termcp, covering security & privacy, agent skills, issue tracker, triage labels and domain docs.
violet AGENTS.md
Instructions for VOD-Studio/violet, covering violet (blog-project), 架构与代码边界, 架构耦合约束, 开发流与命令 (makefile) and 后端 (api/) 须知.
unifi-mcp copilot-instructions.md
Instructions for gordcurrie/unifi-mcp, covering github copilot instructions — unifi-mcp, project overview, architecture, probe any v1 endpoint — substitute path and id as needed and uses a self-signed certificate in this home lab environment.
unifi-mcp AGENTS.md
Instructions for gordcurrie/unifi-mcp, covering unifi-mcp, architecture, environment variables, build and quality gate and network security audit.
yandex-mcp AGENTS.md
Instructions for n-r-w/yandex-mcp, covering project specific rules and information, yandex services supported, tech stack, instructions and golang rules.
libtmux-go AGENTS.md
AGENTS.md instructions for libtmux/libtmux-go, covering agents.md, what is here, which policy applies, change discipline and references.