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 commands/jcarlosrodicio/opencode-agent-orchestration-kit/loopgit clone --depth 1 https://github.com/jcarlosrodicio/opencode-agent-orchestration-kitWhat 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 | $0.00018 | $0.01913 |
| Opus 5 | $0.00009 | $0.00957 |
| Sonnet 5 | $0.00004 | $0.00383 |
| Haiku 4.5 | $0.00002 | $0.00191 |
Grade A, and why
loop 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 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.
How it starts
The opening of the file, as written. The whole thing — 197 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Objective or action:
$ARGUMENTS
Run a manual, resumable engineering loop for work with mechanical validation. This command does not schedule recurring runs or turn the task into unattended automation.
Verifiable invariants
approval_gate: explicit_before_writes
max_iterations_per_invocation: 3
planned_iteration_budget: task_specific_1_to_6
hard_safety_ceiling: 6
completion_authority: reviewer_only
canonical_review_policy: code-review-and-quality/references/review-policy.md
final_review_authority: reviewer
canonical_state_path: .opencode/loops/<slug>.json
history_path: .opencode/loops/<slug>.history.jsonl
lock_path: .opencode/loops/<slug>.lock
human_view_path: .opencode/loops/<slug>.md
worktree_mode: explicit_opt_in
The three-iteration value limits one invocation. The persisted
planned_iteration_budget (planned_iterations in canonical state) is the total budget for the mission and may not
exceed the hard ceiling of six; current_iteration may never exceed that
budget. Once the total budget is exhausted, the loop must stop as blocked or
completed rather than injecting another continuation.
The durable handoff_packet (.opencode/handoffs/<slug>.md with
approval_status) persists human approval to survive session restarts. /loop
keeps .opencode/loops/<slug>.json as canonical state,
.opencode/loops/<slug>.history.jsonl as append-only history, and
.opencode/loops/<slug>.md as the human view. See
docs/ai/harness/agents.md section "Durable /loop state".
Each iteration follows:
developer -> reviewer -> developer (state sync)
Interface
/loop <objective>designs a new loop./loop resume <slug>inspects canonical state before the gate and acquires the durable lock only after explicit approval.- Enable a worktree only when the user explicitly includes
worktreein the arguments or requests it at the approval gate. Otherwise work sequentially in the current checkout. - Every invocation, including a resume, opens a new block of at most three iterations.
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 · 197 lines · 18 tokens per session scan A 658651f40738
loop is a command published in the GitHub repository jcarlosrodicio/opencode-agent-orchestration-kit (105 stars, last pushed 4d ago), licensed Apache-2.0. It adds 18 tokens to every session and 1,913 once invoked, about $0.0001 per session on Opus 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 2026-08-30.
Other commands, from other repositories
execute-parallel
Spawn N concurrent /ai-sdlc execute sessions in tmux panes (max 5). Reads the dispatch-ready frontier, presents top N candidates, asks operator to confirm, then fires one tmux window per task. Coordination via .ai-sdlc/dispatch/sessions/. Resource-gated: refuses if available memory = ncpu. AISDLC-462.
looper
Design and scaffold a Looper agent loop.
loop
The loop pack is dxkit's deterministic preflight/postflight layer for autonomous coding loops — a Claude Code session that keeps working until it decides to stop. It stops a loop from declaring "done" while it has introduced net-new findings, by re-running the guardrail on every Stop and feeding any net-new findings…
selfloop
Start or control a persistent SIPS loop whose only objective is improving SIPS and the agent operating it through measured, verified iterations.
evolution-engine
Scan feedback and generate evolution proposals for rule/skill upgrades.
pack
Pack target repo with repomix. Snapshot for AI context. Cached per featureid.