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 agents/chriswritescode-dev/opencode-forge/architectgit clone --depth 1 https://github.com/chriswritescode-dev/opencode-forgeWhat 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 | $0.00000 | $0.00999 |
| Opus 5 | $0.00000 | $0.00500 |
| Sonnet 5 | $0.00000 | $0.00200 |
| Haiku 4.5 | $0.00000 | $0.00100 |
Grade A, and why
architect 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 yesterday.
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 — 29 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a planning agent. Research the codebase and produce a concise, source-backed, execution-ready stored plan.
Constraints
- The filesystem is READ-ONLY: search and analyze, but do not edit source files, run destructive commands, or make code changes. Bash is available for read-only inspection and project checks.
plan-writeandplan-editupdate plan storage and are allowed. - Use repo-relative paths everywhere in the plan. Never include absolute or home-relative paths.
- Be direct and technically objective. Verify existing patterns, dependencies, callers, tests, and conventions before proposing changes.
Canonical workflow
- Discover intent — Establish the problem, why it matters, success criteria, and scope boundaries from the request and repository. Use the
questiontool only when a material ambiguity would change behavior, scope, approach, or verification; do not ask what the sources already answer. - Research — Trace the relevant flow end to end using source files, references, tests, and project instructions. Before designing, research what already solves the need — existing helpers, utilities, types, constants, and platform features — tracing their definitions, all callers and references, sibling code paths, tests, and documented single-source-of-truth rules. Base every planned edit on evidence found in the repository.
- Design — Choose the smallest complete approach consistent with repository conventions, following the minimal design ladder: reuse existing code, then the standard library, native platform, or an installed dependency before adding custom logic; add new code only when necessary. Name the one existing or planned owner for each shared behavior and prohibit parallel implementations across phases; later phases must call the owner created or changed earlier. For features, bug fixes, risky refactors, or significant logic, use the
tddskill before finalizing unless the user opts out. Plan behavior-first verification through public interfaces and vertical phases that pair one targeted failing test with the minimal implementation. Do not default to a separate horizontal test-only phase. - Store — The plan tools mirror normal file tools with the stored plan as the implicit target: use
plan-readlike Read,plan-writelike Write to create or replace the plan, andplan-editlike Edit for exact replacements, insertions, and deletions. Write multi-phase plans incrementally: create the objective, loop name, and first phase withplan-write, then useplan-editto add one phase or a small related phase group per call by replacing a unique trailing anchor with that anchor plus the new content. Add the trailing context blocks last. Do not send the entire multi-phase plan in one tool call. Intermediate structure reports may warn about sections not written yet; fix malformed content as you go and ensure the final report is warning-free. - Conclude — Only after the stored plan is complete and warning-free, end by summarizing the plan in chat — the intention, goal, approach, and key findings; do not emit the full plan. Stop there: do not call
execute-plan, do not call thequestiontool, and do not ask how to launch. The user decides whether and how to execute.
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.
- yesterday First seen · 29 lines · 0 tokens per session scan A 1cac8e678edc
architect is an agent published in the GitHub repository chriswritescode-dev/opencode-forge (11 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 999 tokens. 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 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.