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/featuregit 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 | $0.00016 | $0.00699 |
| Opus 5 | $0.00008 | $0.00349 |
| Sonnet 5 | $0.00003 | $0.00140 |
| Haiku 4.5 | $0.00002 | $0.00070 |
Grade A, and why
feature 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 3d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Objective:
$ARGUMENTS
Run the feature flow with explicit barriers.
Routing: for free-form messages without a command,
leaduses the declarative routing table indocs/ai/harness/commands.md(section "Declarative routing"). In/featurethe mandatory flow below applies and takes precedence.
Init/context policy
Before the first handoff, confirm only what is needed for correct routing:
cwd: current repository or user-provided target directory.AGENTS.md: applicable local rules when present.git state: clean/dirty state if the change will touch files.validation commands: likely repo checks, without running them yet unless needed to classify.repo docs: local docs that are named or obvious.
Keep this policy lightweight. Small free-form work keeps the fast path; do not turn init into technical discovery.
Mandatory flow
- Analyze the objective only enough to classify scope, uncertainty, and handoffs; do not turn this phase into technical discovery or broad repository/config inspection.
- Apply the base flow:
lead -> designer if applicable -> researcher -> specifier -> developer -> reviewer. - Decide whether design and/or research are needed with minimal routing context. Lead's pre-handoff inspection is limited to obvious signals needed to choose the next agent.
- If the request states or reveals UX, brand, layout, interaction, or visual
acceptance-criteria uncertainty, invoke
designerbefore the first substantive discovery on those topics and wait for handoff. - If the request states or reveals technical/product/API/library/risk/
architecture uncertainty, invoke
researcherbefore the first substantive discovery, broad implementation/config reading, or technical conclusion bylead, then wait for output. - The early handoff to
researcher/designermust include the objective, detected uncertainty, constraints, and expected evidence; it must not pre-resolve the question that motivated delegation. - Do not make
researcheruniversal: if the feature is simple, clear, and has no relevant uncertainty, keep fast routing toward spec/implementation as appropriate. - Parallelize designer and researcher only if their results are independent.
- Synthesize required design/research before invoking
specifier. - Invoke
developeronly for sufficiently specified tasks. - Invoke
reviewerafter implementation with the task/spec, diff base, and original validation evidence. Requirereview_stage: finaland a canonical verdict before closure. - If reviewer returns
needs_changes, route the blocking findings through lead back to developer, then review again. If it returnsblocked, obtain the required evidence or human decision. - Close with changes, validation, and risks.
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.
- 3d ago First seen · 72 lines · 16 tokens per session scan A a34f956be14a
feature is a command published in the GitHub repository jcarlosrodicio/opencode-agent-orchestration-kit (105 stars, last pushed 5d ago), licensed Apache-2.0. It adds 16 tokens to every session and 699 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
prime-pm-role
Prime the PM (project manager) persona for the headless session runner. Receives the user message as $ARGUMENTS.
_prime-rails
These rails apply to every session-runner session (PM, Dev, and Teammate roles). Each role prime instructs you to read and apply this file before acting.
prime-dev-role
Prime the Dev (developer) persona for the headless session runner. No task is present in $ARGUMENTS — the runner will relay the PM's first instruction separately.
prime-teammate-role
Prime the Teammate persona for the headless session runner. Receives the user message as $ARGUMENTS.
update
Bring this machine to latest main — sync deps, verify environment, restart services. Runs deterministically via inline bash execution, not model discretion.
release
This compatibility command keeps /oh-my-claudecode:release available without loading the full release skill description in every Claude Code session.