Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/ViryaZheng/recomby-geonpx agentmods add commands/viryazheng/recomby-geo/04-content-briefWrote 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/viryazheng/recomby-geo/04-content-brief)<a href="https://agentmods.dev/commands/viryazheng/recomby-geo/04-content-brief"><img src="https://agentmods.dev/badge/commands/viryazheng/recomby-geo/04-content-brief/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/viryazheng/recomby-geo/04-content-brief"><img src="https://agentmods.dev/badge/commands/viryazheng/recomby-geo/04-content-brief.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00096 | $0.02828 |
| Opus 5 | $0.00048 | $0.01414 |
| Sonnet 5 | $0.00019 | $0.00566 |
| Haiku 4.5 | $0.00010 | $0.00283 |
Grade A, and why
04-content-brief 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 11d 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 — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.
04 · Content Brief — Human-in-Loop Checkpoint
This is the most important command in the pipeline. Without it, the system collapses into the same generic AI-content-mill that has saturated the GEO commodity layer. With it, the system produces content AI engines actually cite — because it contains things AI can't auto-generate.
The principle: the pipeline builds the scaffold; humans fill the soul. This command outputs scaffold with explicit, named, schema-tight blank spaces. Until those blanks are filled, 05-production refuses to run.
Inputs
clients/<slug>/brand_context.jsonclients/<slug>/content_priorities.json- A specific
priority_id(passed by user or auto-picked: highest unbriefed).
Output
clients/<slug>/briefs/<priority-id>.md— human-readable brief with REQUIRED-FILL slots clearly marked.clients/<slug>/briefs/<priority-id>.meta.json— machine-readable metadata: which slots are filled, by whom, when.clients/<slug>/briefs/<priority-id>.html— interactive fill-form for the client/expert (rendered by thegeo-review-htmlskill; this is the artifact the reviewer actually opens, not the.md).
Procedure
Step 1 — Load context
jq --arg id "<priority-id>" '.priorities[] | select(.id==$id)' \
clients/<slug>/content_priorities.json
Pull the priority record. Cross-reference brand_context for the relevant ICP, voice samples, and competitor positioning.
Step 2 — Choose the angle
Generate 2–3 candidate angles for the priority. An angle is a one-sentence statement of the unique thesis the piece will defend. Bad angle: "Best CRM for small business." Good angle: "Why bookkeepers (not sales teams) should own CRM choice in firms under 20 people, based on our 200-firm dataset."
The angle MUST satisfy:
- Aligns with
core_differentiatorandcompetitive_moatfrom brand_context. - Targets the priority's
intentdirectly. - Cannot be written without the brand's specific expertise/data/perspective (i.e., GPT-4 with a Google search couldn't produce it).
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.
- 11d ago First seen · 300 lines · 96 tokens per session scan A 784f8d8d01eb
04-content-brief is a command published in the GitHub repository ViryaZheng/recomby-geo (454 stars, last pushed 2mo ago), licensed MIT. It adds 96 tokens to every session and 2,828 once invoked, about $0.0005 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
commit-all
Stage and commit all current changes with a single message. Auto Git Push will pick up and push to origin within 30s.
check
Run the full local quality gate — TypeScript check, build, dependency audit (prod-only), skill registry validate. Fast pre-commit / pre-suggestdeploy sweep.
registry
Refresh the skills registry — regenerate SHA-256 manifest from disk, then validate. Run after editing any file under .agents/skills/.
onboard
/anty:onboard — QUEST-Based Conversational Interview.
scan
/anty:scan — Analysis Cycle (10-Point).
quickstart
/anty:quickstart — 5-Minute Quick Onboarding to First Action.