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 skills add matteoroversi/lookward --skill lw-forgegit clone --depth 1 https://github.com/matteoroversi/lookwardWrote 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/skills/matteoroversi/lookward/lw-forge)<a href="https://agentmods.dev/skills/matteoroversi/lookward/lw-forge"><img src="https://agentmods.dev/badge/skills/matteoroversi/lookward/lw-forge/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/skills/matteoroversi/lookward/lw-forge"><img src="https://agentmods.dev/badge/skills/matteoroversi/lookward/lw-forge.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 | $0.00065 | $0.01422 |
| Opus 5.5 | $0.00026 | $0.00569 |
| Sonnet 5.5 | $0.00013 | $0.00284 |
| Haiku 4.5 | $0.00006 | $0.00142 |
Grade A, and why
lw-forge 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 2mo 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
lw-forge
The factory — it agentifies the kinetic layer. Once the ontology means something, lw-forge turns its actions into skills/agents/workflows and its actors into the agents that run them. It reads the layers — semantic (objects + links), kinetic (actions + actors), dynamic (governance) — and the will on top (intent + direction), and proposes bespoke tooling fitted to this world model. Loads ontology/contract.md. The base set stays minimal; the surface grows by forging. Everything forged is proposed (code + manifest) for human review before it runs.
What it forges
- The kinetic-execution family — skills, agents, and workflows (the three execution primitives). This is the payoff: each
actioninontology.mdbecomes an executable. An action is already a typed contract in the contract —parameters+rules(the closed verb catalog) +submission_criteria+side_effects+execution: auto | confirm.lw-forgecompiles it:
- Skills (the primary output) — one action the subject repeats (prep, write, brief, assess…). Output: a new
skills/<name>/SKILL.md. The action'sexecution: auto | confirmflag carries over, so the same definition serves a human invoking it and an agent calling it as a tool — governed identically. - Agents — bounded, independent, or parallel workers a skill spawns (a reviewer, a researcher, a verifier), or the agentified form of an actor in the ontology. Forged when a skill needs silent heavy lifting it shouldn't do in the user's conversation.
- Workflows — deterministic orchestrations of skills/agents for a recurring multi-step job (a sequence of actions along the will's loop). Forge proposes one when the signal shows a repeated sequence (A→B→C) or a fan-out worth running deterministically — only when volume/recurrence demands it, never by default. Propose-don't-apply holds: the workflow is reviewed before it runs, and any write goes through the action write-back (append-only to
edits/, human-gated).
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.
- 2mo ago First seen · 45 lines · 65 tokens per session scan A d765dc9260e4
lw-forge is a skill published in the GitHub repository matteoroversi/lookward (7 stars, last pushed 4mo ago), licensed MIT. It adds 65 tokens to every session and 1,422 once invoked, about $0.0003 per session on Opus 5.5. 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.
lw-forge compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| lw-forgematteoroversi/lookward | Skill | 7 | 4mo ago | MIT | A | 65 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |