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/joaoequer/oficina/initgit clone --depth 1 https://github.com/JoaoEquer/OficinaWhat 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 | $0.00032 | $0.01158 |
| Opus 5 | $0.00016 | $0.00579 |
| Sonnet 5 | $0.00006 | $0.00232 |
| Haiku 4.5 | $0.00003 | $0.00116 |
Grade A, and why
init 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Configure the current project with the Oficina harness (https://github.com/JoaoEquer/Oficina). Your goal: by the end, any AI (Claude Code, Gemini, Cursor, Codex) opened in this project works in the house pattern automatically.
Step 1 — Understand the project BEFORE generating anything
Investigate the current repository:
- Read
package.json,README.mdand the folder structure - Identify: project name, real stack (NestJS? Prisma? which database?), whether a CRUD module already exists to serve as the mold, and where the data model document lives (look in
docs/) - If the user passed a description in
$ARGUMENTS, use it as primary context
If you cannot identify the essentials (name, purpose, whether it is multi-tenant), ask the user — at most 3 objective questions. Do not invent context.
Step 2 — Obtain the rules content
Read the harness's four rule files. Try in this order:
- From the installed plugin directory (you are running from inside it):
rules/engineering-philosophy.md,rules/git-workflow.md,rules/security-baseline.md,rules/agent-workflow.md - If not found, fetch from the public repository:
- https://raw.githubusercontent.com/JoaoEquer/Oficina/main/rules/engineering-philosophy.md
- https://raw.githubusercontent.com/JoaoEquer/Oficina/main/rules/git-workflow.md
- https://raw.githubusercontent.com/JoaoEquer/Oficina/main/rules/security-baseline.md
- https://raw.githubusercontent.com/JoaoEquer/Oficina/main/rules/agent-workflow.md
Step 2.5 — Check for a workspace-level harness
Look for an AGENTS.md in the parent directories (e.g. a workspace folder like Wibi/AGENTS.md) that already embeds the Oficina rules. If one exists, do not re-embed the rules in the project's AGENTS.md — harnesses load parent context files hierarchically, and duplicating them wastes context on every turn. In that case the project file contains only: context, source of truth, locked decisions and pending items, plus one line: "Workspace-level rules inherited from the parent AGENTS.md (Oficina harness)."
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 · 93 lines · 32 tokens per session scan A 01e696e47387
init is a command published in the GitHub repository JoaoEquer/Oficina (2 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 1,158 once invoked, about $0.0002 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 2026-08-31.
Other commands, from other repositories
product-spec-builder
Gather requirements and generate/update Product-Spec.md through structured interview.
domain-mapper
Research and map a domain/industry/technology into a structured Markdown database.
change-manager
Brownfield change workflow — propose, apply, verify, archive (OpenSpec-aligned).
code-review
Run parallel specialized review agents and produce aggregated quality report.
reqforge-greenkeeper
Diagnose and fix ReqForge repository release gate failures.
bug-fixer
Diagnose root cause and fix bugs through systematic debugging.