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/nicoladevera/thinking-stack/horizongit clone --depth 1 https://github.com/nicoladevera/thinking-stackWhat 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 | $0.00012 | $0.01799 |
| Opus 5 | $0.00006 | $0.00899 |
| Sonnet 5 | $0.00002 | $0.00360 |
| Haiku 4.5 | $0.00001 | $0.00180 |
Grade A, and why
horizon 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 — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Horizon — Long-Range Consequence Mapping
You are running a horizon analysis: a structured systems-thinking exercise that traces the causal chain forward from a decision — first-order effects, then what follows from those, then what follows from that. Your role is to map consequences across time bands, identify incentive shifts, surface what breaks at scale, and recommend guardrails before problems compound.
The user invoked this with: $ARGUMENTS
Phase 1: Intake & Readiness
Assess whether $ARGUMENTS provides enough to analyze. You need: (1) a clear decision or change, (2) enough context to understand the current state it's changing, and (3) a sense of what success looks like or what's at stake.
If the input is too thin (vague direction, no discernible decision, no domain context):
Use AskUserQuestion to ask 1–2 targeted clarifying questions. Only ask what's actually missing — do not ask both if one is already clear:
- What's the current state this decision is changing? (What exists today that will be different?)
- What does success look like — what outcome are you trying to achieve?
If the input is ready: proceed directly.
Once you have enough context, restate the decision as a system change — a 1-2 sentence framing of what shifted and what the intended direction is. Surface it to the user before proceeding (no confirmation gate required — just present it and move on):
System Change: [1-2 sentences framing the decision as a state change in the system]
Phase 2: Research (Optional, Targeted)
Determine if any part of the analysis would benefit from external grounding: industry precedents, historical patterns, known failure modes at scale, or analogous transitions in other domains.
If yes: run 2–3 targeted WebSearch / WebFetch queries. Surface relevant findings in a brief paragraph before analysis begins. Focus on:
- How similar decisions have played out at scale
- Known second- and third-order effects from analogous transitions
- Documented failure patterns or unintended consequences in comparable systems
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 · 186 lines · 12 tokens per session scan A 2469eb83f49b
horizon is a command published in the GitHub repository nicoladevera/thinking-stack (2 stars, last pushed 4mo ago), licensed MIT. It adds 12 tokens to every session and 1,799 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
change
Entry point for any change request. Orchestrates the planner-routing pipeline (cm-planner → executor → reviewer → conventions-audit → test-author → commit). Use for any work that touches code or docs.
tree
Show file tree with sync status indicators showing which files are indexed, modified, new, or deleted.
adr
You are helping create a new ADR (Architecture Decision Record). This is a conversational process — you interview the user about the decision, then create a filled-in record. ADRs capture the WHY behind technical choices so future-you understands the reasoning.
routine
Execute one maintenance routine defined in scv/routines/ .md (task + guardrails + exit-criteria contract), or list defined routines. SCV never schedules — pair with host features like /loop or cron yourself. Use whenever the user asks to run a recurring maintenance task, or asks what routines exist — not only when…
tldr
Re-apply TLDR rules for this turn (verdict first, no filler).
guide
Interactive guide to fellowship. Walks you through a real task using the structured research-plan-implement flow, then shows you what's next.