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 skills/slhaf/agenticgpt/skill-installernpx skills add slhaf/AgenticGPT --skill skill-installergit clone --depth 1 https://github.com/slhaf/AgenticGPTWrote 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/slhaf/agenticgpt/skill-installer)<a href="https://agentmods.dev/skills/slhaf/agenticgpt/skill-installer"><img src="https://agentmods.dev/badge/skills/slhaf/agenticgpt/skill-installer.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 | $0.00017 | $0.00486 |
| Opus 5 | $0.00009 | $0.00243 |
| Sonnet 5 | $0.00003 | $0.00097 |
| Haiku 4.5 | $0.00002 | $0.00049 |
Grade A, and why
skill-installer 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 5d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill installer
Use this skill to guide installation of another skill in the active Room Agent.
Skills are managed by the Room Agent, so these tools do not take an agentId.
Recommended workflow
- Inspect the available skills with
skills.listor read this guide withskills.read. - Start one installation with
skills.install. Always provide the targetidexplicitly. Use a structured public GitHub source when installing from a repository, or use thefilessource for URL-backed and inline files. - Treat the immediate response as an acknowledgement. Keep its
installIdand pollskills.install.get; status lookup waits up to five seconds by default and acceptswaitSecondsfrom 0 through 30. - Use
skills.install.cancelwhen queued or cancellable work should stop. Cancellation is cooperative and cannot interrupt an atomic commit. - After completion, call
skills.readto inspectSKILL.mdor a safe package-relative resource. A newly installed skill is active by default.
Replacement and retry rules
Existing skills are not replaced unless replaceExisting: true is explicit.
Replacement archives the old package under skills/.archive so it can be
restored if the commit fails. An optional Room-scoped idempotencyKey makes a
fresh retry return the original installation instead of creating a duplicate.
Only public GitHub repositories and public HTTPS file URLs are supported in
the initial version. Do not put credentials in URLs. Inline content and
contentBase64 remain available for generated files. Installation validates
the package and never executes its scripts.
Running installed scripts
For an active workspace-backed skill, skills.run accepts the skill id, a
package-relative executable path below scripts/, optional direct args, an
optional policy-validated workingDirectory, and a bounded waitSeconds.
It returns a managed Job envelope. Fast terminal executions are included in
the response; otherwise use job.get with bounded waitSeconds and
job.cancel with that jobId.
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.
- 5d ago First seen · 51 lines · 17 tokens per session scan A 1101b3088f3f
skill-installer is a skill published in the GitHub repository slhaf/AgenticGPT (5 stars, last pushed 10d ago), licensed MIT. It adds 17 tokens to every session and 486 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 skills, from other repositories
bugs
Proactively sweep a Pake area for latent UX and runtime defects before users report them, using this repo's own fix history and boundary archetypes, and confirm every finding with a probe run this turn. Use for 找找有没有bug, 主动找bug, 有没有隐患, 排查隐患, 举一反三, 上线前扫一遍, 这块可靠吗, latent bug scan, sibling sweep after a fix. Not for an…
release
Prepare, validate, and publish a Pake release. Not for version bumps without release intent.
pake
Package any website or local web build into a lightweight desktop app using Pake (Tauri/Rust). Use when the user wants to: wrap a URL as a native app, build a desktop app from a website or a local dist/ folder, use Pake CLI to package a page, set up proxy for a packaged app, customize app icons or bundle IDs, or…
github-ops
GitHub issue, PR, and release operations via gh CLI. Not for code review or release builds.
starknet-agentic-skills
Routes Starknet agent, wallet, DeFi, identity, SDK, and Cairo contract work to the smallest focused skill module.
awesome-git-history-rebuild
Erases a repository's git history and rebuilds it as a curated commit series over the same tree: a split plan the user approves (proposed here, or supplied by awesome-git-commit-plan and validated first), the repo's own commit rules obeyed (commitlint, hooks, CONTRIBUTING), paced timestamps anchorable to the earliest…