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/glbforge/glbforgenpx agentmods add skills/glbforge/glbforge/glbforge-info-passWrote 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/glbforge/glbforge/glbforge-info-pass)<a href="https://agentmods.dev/skills/glbforge/glbforge/glbforge-info-pass"><img src="https://agentmods.dev/badge/skills/glbforge/glbforge/glbforge-info-pass/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/glbforge/glbforge/glbforge-info-pass"><img src="https://agentmods.dev/badge/skills/glbforge/glbforge/glbforge-info-pass.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 | $0.00080 | $0.01727 |
| Opus 5.5 | $0.00032 | $0.00691 |
| Sonnet 5.5 | $0.00016 | $0.00345 |
| Haiku 4.5 | $0.00008 | $0.00173 |
Grade A, and why
glbforge-info-pass 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 1mo 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
One GLBForge info pass
GLBForge is read before it is run. A person reads glbforge.dev; an agent reads
/llms.txt and acts on it without checking. So the information surface is part
of the product, and a stale claim there is a defect with a blast radius — an
agent that believes a command exists writes a plan around it.
One pass, then stop. The question behind every decision: would someone who only read this end up with the right expectation?
Rules for the whole pass
- Open a PR. Never merge, never deploy, never tag. Publishing the site is
wrangler deployfrom the maintainer's machine;scripts/live-check.mjsalready notices when the deploy is behindorigin/main, so a merged PR cannot be quietly forgotten. Taggingv*triggers the npm release. - No paid path, same as the agent loop. Read-only HTTP to glbforge.dev is fine.
- Say only what you checked. This repo's own rule is that a message states what was measured. A PR that claims a page is accurate because it looks accurate is the drift it was supposed to fix.
- A pass that finds nothing opens no PR. Say so and stop.
0. Nothing else is already on it
Run this command. Not a PR-listing tool — gh filters server-side and
returns three fields, where a generic "list pull requests" call returns every
open PR with its entire body. This repo keeps fifteen-odd agent-loop PRs open
with long descriptions: the 2026-09-27 run's call came back at 150KB, blew the
token limit, and cost six turns of grep and python3 against a saved dump
to recover a list of branch names.
gh pr list --state open --json number,title,headRefName \
--jq '.[] | select(.headRefName | startswith("info-pass/") or startswith("agent-loop/"))'
Anything open is claimed — including an agent-loop PR that already touches the docs for the feature it shipped. Do not open a rival.
The same reflex applies for the rest of the pass: reach for gh or git with
a filter or a format flag before a tool that returns whole objects. You are
reading this repo to find one difference, not to load 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.
- 1mo ago Changed · +11 lines 547b70e58a0b
- 1mo ago First seen · 132 lines · 80 tokens per session scan A dd3e2c16faf9
glbforge-info-pass is a skill published in the GitHub repository glbforge/glbforge (3 stars, last pushed 29d ago), licensed MIT. It adds 80 tokens to every session and 1,727 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 2026-09-26.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…