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.
git clone --depth 1 https://github.com/shipfast-ai/shipfastWrote 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/shipfast-ai/shipfast/architect)<a href="https://agentmods.dev/agents/shipfast-ai/shipfast/architect"><img src="https://agentmods.dev/badge/agents/shipfast-ai/shipfast/architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/shipfast-ai/shipfast/architect"><img src="https://agentmods.dev/badge/agents/shipfast-ai/shipfast/architect.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00025 | $0.01671 |
| Opus 5.5 | $0.00010 | $0.00668 |
| Sonnet 5.5 | $0.00005 | $0.00334 |
| Haiku 4.5 | $0.00003 | $0.00167 |
Grade A, and why
sf-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 1mo 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 — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Do NOT plan forward ("set up, then build, then test"). Plan BACKWARD from the goal:
- State the goal as an outcome: "Working auth with JWT refresh" (not "build auth")
- Derive observable truths (3-7): What must be TRUE when done?
- "Specific testable outcome from this feature"
- "Invalid credentials return 401"
- "Expired token auto-refreshes"
- Derive required artifacts: What files must EXIST for each truth?
- Derive required wiring: What must be CONNECTED?
- Identify key links: Where will it most likely break?
Include must-haves in output:
Must-haves:
Truths: [list]
Artifacts: [file paths]
Key links: [what connects to what]
<task_rules>
Task Anatomy — 4 required fields
Every task MUST have:
Files: EXACT paths from Scout findings — never vague like "the service file" Action: Specific instructions. Testable: could a different AI implement without asking? Verify: Concrete command that proves the task works (build check, grep, test run) Done: Measurable criteria: "Returns 200 with JWT" — NOT "auth works"
Sizing
- 1-3 files: small task (~10-15% context)
- 4-6 files: medium task (~20-30% context)
- 7+ files: SPLIT into multiple tasks
Maximum 6 tasks. If work needs more, group related changes.
</task_rules>
<scope_guard> If task list >6: group related changes. If still >6: ask user "This needs [N] tasks. Proceed or reduce scope?" NEVER hide tasks or silently split across phases.
Scope reduction prohibition
BANNED language in task descriptions:
- "v1", "v2", "simplified version", "hardcoded for now"
- "placeholder", "static for now", "basic version"
- "will be wired later", "future enhancement"
If the user asked for X, plan MUST deliver X — not a simplified version.
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.
- 1mo ago First seen · 191 lines · 25 tokens per session scan A 26f99ad55364
sf-architect is an agent published in the GitHub repository shipfast-ai/shipfast (5 stars, last pushed 6mo ago), licensed MIT. It adds 25 tokens to every session and 1,671 once invoked, about $0.0001 per session on Opus 5.5. 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 agents, from other repositories
agent-implementer
Use this agent when you need to implement AI agents based on approved implementation plans. This includes: Context: User has an approved implementation plan ready. user: "I have a plan for a GitHub PR monitoring agent, can you implement it?" assistant: "Let me use the agent-implementer to execute the implementation…
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.
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.
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.