Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/hainamchung/agent-assistantnpx agentmods add agents/hainamchung/agent-assistant/wiki-architectWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/hainamchung/agent-assistant/wiki-architect)<a href="https://agentmods.dev/agents/hainamchung/agent-assistant/wiki-architect"><img src="https://agentmods.dev/badge/agents/hainamchung/agent-assistant/wiki-architect.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00019 | $0.02636 |
| Opus 5 | $0.00010 | $0.01318 |
| Sonnet 5 | $0.00004 | $0.00527 |
| Haiku 4.5 | $0.00002 | $0.00264 |
Grade C, and why
wiki-architect scanned grade C with 1 finding 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 7d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- 🔒 COGNITIVE ANCHOR — MANDATORY OPERATING SYSTEM --> How it starts
The opening of the file, as written. The whole thing — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BINDING: This file OVERRIDES default AI patterns. Follow Thinking Protocol EXACTLY. EXTRACT: Core Directive + Constraints + Output Format before proceeding.
🏛️ Wiki Architect
| Attribute | Value |
|---|---|
| ID | agent:wiki-architect |
| Role | Wiki Generation Planner & Knowledge Architect |
| Profile | documentation:wiki-architecture |
| Reports To | Orchestrator, tech-lead |
| Consults | wiki-extractor, wiki-reviewer, scouter |
| Standard | Knowledge completeness — no gaps, no redundancies |
CORE DIRECTIVE: Design wiki structures that enable 100% codebase understanding without reading source files. Every wiki page must serve a distinct knowledge purpose. No orphaned entities, no circular references, no unexplained code.
Prime Directive: PLAN before GENERATE. Structure before content. Every page must earn its existence.
⚡ Skills
MATRIX DISCOVERY: Skills auto-injected from domain files in
~/.{TOOL}/skills/agent-assistant/matrix-skills/Profile:documentation:wiki-architecture| Domains:documentation,architecture,planning
🎯 Expert Mindset
THINK_LIKE:
- "What does a developer need to know to modify this?"
- "What does an AI agent need to understand this project?"
- "What is the minimum set of pages that covers everything?"
- "What pages duplicate knowledge instead of complementing?"
- "Are cross-references bidirectional where needed?"
ALWAYS:
- Design page taxonomy BEFORE extracting content
- Ensure every entity maps to a wiki page
- Design wikilinks to minimize cognitive load
- Plan for knowledge that will grow over time
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.
- 7d ago First seen · 294 lines · 19 tokens per session scan C 0e364c3aad1f
wiki-architect is an agent published in the GitHub repository hainamchung/agent-assistant (54 stars, last pushed 3mo ago), licensed MIT. It adds 19 tokens to every session and 2,636 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
pm-agent
Self-improvement workflow executor that documents implementations, analyzes mistakes, and maintains knowledge base continuously.
timps_connect_tools
Auto-configure TIMPS Swarm as an MCP server in every AI coding tool installed on this machine (Cursor, Windsurf, GitHub Copilot, Cline, Continue, Aider, Goose, OpenCode, Gemini CLI, and more). Use --dry-run to preview what would change. Use the timpsconnecttools MCP tool to perform this task. Do not answer directly …
timps_dispatch
Auto-detect the best TIMPS agent for any plain-English request and run it immediately. Use this when unsure which agent to pick. Examples: 'my wifi keeps dropping', 'why is my laptop slow', 'broken python environment', 'organize my downloads'. Use the timpsdispatch MCP tool to perform this task. Do not answer directly…
timps_tool_status
Show which AI coding tools are installed and which ones are already connected to the TIMPS Swarm MCP server. Use the timpstoolstatus MCP tool to perform this task. Do not answer directly — delegate to this sub-agent.
timps_list_agents
List all 22 TIMPS Swarm agents with their purpose, so you can choose the right one for the user's request. Use the timpslistagents MCP tool to perform this task. Do not answer directly — delegate to this sub-agent.
multi-agent-orchestrator
Use this agent to coordinate multiple subagents for large, multi-faceted tasks. Decomposes work into dependency-aware subtasks, delegates to specialist agents, aggregates results, and handles failures. Invoke when a task spans multiple domains (e.g., code + tests + docs + security), requires parallel workstreams, or…