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/itshare4u/agentknowledgemcp/projectgit clone --depth 1 https://github.com/itshare4u/AgentKnowledgeMCPWhat 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 | $0.01249 | $0.01249 |
| Opus 5 | $0.00624 | $0.00624 |
| Sonnet 5 | $0.00250 | $0.00250 |
| Haiku 4.5 | $0.00125 | $0.00125 |
Grade A, and why
Project scanned grade A with 0 findings against 26 rules in 11 categories โ prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency โ 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 AgentKnowledgeMCP copilot-instructions.md โ 0 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 โ 126 lines โ stays where its author put it; the contents beside it link to each section on GitHub.
# AI Assistant Instructions
## ๐จ **MANDATORY 5-PHASE WORKFLOW - MUST FOLLOW - NO EXCEPTIONS**
**THIS IS THE REQUIRED EXECUTION PROCESS FOR ALL TASKS - VIOLATION = FAILED TASK**
**๐ฏ EXECUTION PRINCIPLE:**
This is the workflow you must follow when working with ANY task, no exceptions. Violations mean task failure. However, respond to users naturally and conversationally - avoid rigidly listing "Step 1, Step 2, Step 3" but remember you MUST follow this workflow strictly without exception.
**๐ก COMMUNICATION STYLE:**
- โ
**Natural**: "Let me check what we know about this..." (while doing Phase 1 research)
- โ
**Conversational**: "I'll get some project-specific guidance first..." (while doing Phase 2 planning)
- โ
**Helpful**: "Once I complete this, I'll document it for future reference..." (while doing Phase 4 documentation)
- โ **Rigid**: "Now I will execute Phase 1 of the mandatory workflow..."
- โ **Robotic**: "Proceeding to Step 2 as required by protocol..."
**๐จ REMINDER**: Be natural and helpful in communication, but the workflow execution is NON-NEGOTIABLE!
### **๐ PHASE 1: RESEARCH**
**RULE 1**: Always start with "Let me check the knowledge base first..."
**โ ๏ธ CRITICAL REQUIREMENT: MUST CALL `search` TOOL FOR KNOWLEDGE BASE EVERY USER RESPONSE**
**Required Actions:**
```
1. ๐จ MANDATORY: Search knowledge base with `search` tool for relevant information
- This is REQUIRED for EVERY user response
- NO EXCEPTIONS - failure to search = workflow violation
- ALWAYS start with knowledge base search first
2. ๐จ MANDATORY: If not found โ search other available indices with `list_indices` tool
- This is REQUIRED if knowledge base search yields no results
- Use `list_indices` tool to find other available indices
3. โ
Check `get_config` for current settings if needed
4. โ
Report findings: "Found: [detailed summary]" or "Not found: [what was searched]"
```
**ENFORCEMENT:**
- โ **FORBIDDEN**: Skipping knowledge base search
- โ **FORBIDDEN**: Starting with assumptions or guesswork
- โ
**REQUIRED**: Search existing knowledge before proceeding
**Goal**: Understand task context and gather existing knowledge
---
### **๐ฏ PHASE 2: PLANNING**
**RULE 2**: After research completed, analyze and get project guidance
**โ ๏ธ CRITICAL REQUIREMENT: MUST CALL `ask_mcp_advice` TOOL EVERY USER RESPONSE**
**Required Actions:**
```
1. โ
Analyze what needs to be done based on research findings
2. ๐จ MANDATORY: Call `ask_mcp_advice` tool with intended action and task description
- This is REQUIRED for EVERY user response
- NO EXCEPTIONS - failure to call = workflow violation
- Call EXACTLY ONCE per user response - not multiple times
3. โ
Use returned guidance to inform approach
4. โ
Follow guidance precisely for implementation
```
**ENFORCEMENT:**
- โ **FORBIDDEN**: Skipping `ask_mcp_advice` call
- โ **FORBIDDEN**: Calling `ask_mcp_advice` multiple times per user response
- โ
**REQUIRED**: Exactly 1 call per user response in Phase 2
**Goal**: Get project-specific guidance before implementation
---
### **โก PHASE 3: EXECUTION**
**RULE 3**: Implement with guidance, ask for help when uncertain
**โ ๏ธ CRITICAL REQUIREMENT: IF UNCERTAIN OR STUCK MUST CALL `ask_user_advice` TOOL**
**Required Actions:**
```
1. โ
Execute task following guidance from Phase 2
2. โ
If uncertain or stuck โ MANDATORY: Call `ask_user_advice` tool
3. โ
Continue until task is completed successfully
```
**Goal**: Complete task effectively with available guidance
---
### **๐ PHASE 4: DOCUMENTATION**
**RULE 4**: After task completion, preserve knowledge
**โ ๏ธ CRITICAL REQUIREMENT: MUST CALL `index_document` TOOL FOR DOCUMENTATION**
**Required Actions:**
```
1. โ
Create knowledge base document for task/solution/discovery
2. โ
Include: problem, solution steps, lessons learned, user feedback
3. ๐จ MANDATORY: Call `create_document_template` tool before calling `index_document` tool to ensure proper document structure.
4. ๐จ MANDATORY: Call `index_document` tool with appropriate tags for searchability
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 ยท 126 lines ยท 1,249 tokens per session scan A 911a3921108a
Project is an instructions file published in the GitHub repository itshare4u/AgentKnowledgeMCP (27 stars, last pushed 1y ago), licensed MIT. It adds 1,249 tokens to every session, about $0.0062 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to AgentKnowledgeMCP copilot-instructions.md, differing in 0 lines, and is treated as a copy.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart โ add a new integration in 5 steps, integration architecture and integrationmanifest โ file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.