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/Gekkos-tech/agency-osWrote 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/commands/gekkos-tech/agency-os/proposal)<a href="https://agentmods.dev/commands/gekkos-tech/agency-os/proposal"><img src="https://agentmods.dev/badge/commands/gekkos-tech/agency-os/proposal.svg" alt="Measured on agentmods" 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 | $0.00018 | $0.00347 |
| Opus 5 | $0.00009 | $0.00173 |
| Sonnet 5 | $0.00004 | $0.00069 |
| Haiku 4.5 | $0.00002 | $0.00035 |
Grade A, and why
proposal 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 7d 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.
What it actually says
Draft a client proposal. Additional context: $ARGUMENTS
- Read
docs/brief.md(anddocs/concept.mdif present). If no brief exists, run a compressed BRIEF phase first via theagency-workflowskill — a proposal without a brief is guesswork. - Produce
docs/proposal.mdin the client's language:- Ausgangslage / Situation — the client's goal in their words
- Lösung / Approach — what will be built and why it serves the goal
- Leistungen / Deliverables — per phase (Konzept, Design, Umsetzung, QA, Launch), each with concrete outputs the client can verify
- Ablauf & Timeline — phases with durations and client touchpoints (approvals needed, content delivery deadlines)
- Mitwirkung des Kunden — what the client must provide and when
- Annahmen & Abgrenzung — explicit assumptions and out-of-scope list (this section prevents scope creep; make it specific)
- Investition — pricing per phase if hints were given; otherwise leave a clearly marked placeholder structure — never invent prices
- Keep it honest and concrete: no superlatives, no vague "premium quality" claims — name the actual deliverables (e.g. "design reference image per section", "QA report", "launch checklist").
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.
- 7d ago First seen · 26 lines · 18 tokens per session scan A 831f441a34dc
proposal is a command published in the GitHub repository Gekkos-tech/agency-os (5 stars, last pushed 1mo ago), licensed MIT. It adds 18 tokens to every session and 347 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 2026-08-31.
Other commands, from other repositories
agent-build-feature
A strict brief that makes a coding agent implement a feature end to end, matching the codebase and verifying it works.
agent-system-prompt
Write a system prompt for an AI agent that defines its role, tools, boundaries, and output clearly enough to act reliably.
estimate-work
Estimate a task or project honestly, by decomposing it, surfacing unknowns, and giving a range not a false point.
orchestrator-prompt
Write an orchestrator prompt that coordinates worker agents: delegating, tracking, and synthesizing their results.
bug-report-triage
Turn a vague bug report into a structured, actionable ticket with severity and next steps.
cold-email
Write a short, specific cold outreach email that respects the reader's time and gives a real reason to reply.