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/abhiyoheswaran1/agentloopkit/opencodegit clone --depth 1 https://github.com/abhiyoheswaran1/AgentLoopKitWhat 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 | $0.00209 |
| Opus 5 | $0.00000 | $0.00105 |
| Sonnet 5 | $0.00000 | $0.00042 |
| Haiku 4.5 | $0.00000 | $0.00021 |
Grade A, and why
opencode 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.
What it actually says
OpenCode Agent Instructions
Use AgentLoopKit as the working loop for this repository.
Before editing:
- Read AGENTS.md.
- Read AGENTLOOP.md.
- Check .agentloop/tasks/.
- Run
agentloop doctor --redact-pathsbefore Start when you need to confirm agent-readiness guidance is current. - Before broad file reads or long-session continuation, use one preflight: call MCP tool
agentloop_startwhen MCP tools are configured; otherwise runagentloop start --for generic --goal implement --redact-paths. Useagentloop context handlesto list available handles, thenagentloop context show <handle>to expand local source truth only when needed and avoid broad repo reads.
Rules:
- Do not take destructive actions without explicit approval.
- Prefer small diffs.
- Verify before completion.
- Summarize changed files, tests, risks, and rollback notes.
If OpenCode config conventions differ in your setup, paste or link this file into the appropriate agent instructions.
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 · 21 lines · 0 tokens per session scan A 25208e0f1620
opencode is an agent published in the GitHub repository abhiyoheswaran1/AgentLoopKit (1 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 209 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 2026-08-31.
Other agents, from other repositories
Jump Start: Analyst
Phase 1 agent focused on user research, personas, journeys, and value proposition development.
Jump Start: Architect
Phase 3 agent responsible for translating the PRD into a technical blueprint and an ordered implementation plan. This includes technology selection, component design, data modeling, API specification, and documenting architectural decisions.
developer
You are The Developer, the Phase 4 agent in the Jump Start framework. Your role is to execute the implementation plan produced by the Architect, writing code that faithfully implements the specifications. You are methodical, test-driven, and disciplined. You follow the plan, write clean code, and verify your work…
Jump Start: Product Manager
Phase 2 -- Translate the product concept into a detailed, unambiguous Product Requirements Document (PRD) with user stories, acceptance criteria, and prioritization.
challenger
You are The Challenger, the Phase 0 agent in the Jump Start framework. Your role is to rigorously interrogate the human's initial idea, assumption, or problem statement before any product thinking, planning, or building begins.
Jump Start: Developer
Phase 4 -- Execute the implementation plan task by task, writing tested code.