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 KaydenClark/LLM_Workbench --skill make-it-sogit clone --depth 1 https://github.com/KaydenClark/LLM_WorkbenchWrote 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/kaydenclark/llm_workbench/make-it-so)<a href="https://agentmods.dev/skills/kaydenclark/llm_workbench/make-it-so"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/make-it-so/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/kaydenclark/llm_workbench/make-it-so"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/make-it-so.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 | $0.00038 | $0.00481 |
| Opus 5 | $0.00019 | $0.00241 |
| Sonnet 5 | $0.00008 | $0.00096 |
| Haiku 4.5 | $0.00004 | $0.00048 |
Grade A, and why
make-it-so 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 today.
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.
What it actually says
Resolve the particular settled work and the user's endpoint before composing helpers. The current request controls every step. A specification-only, promotion-only or handoff-only request remains exactly that scope, even when it names this skill. Never treat invocation as approval of unrelated pending choices or as permission supplied by a notepad. Without a narrower endpoint, explicit approval to build includes the agreed implementation and authorized recovery under the project Contract; main publication stays owner-controlled.
Read the Contract and workbench/manifest.json. Compose notepad to resume the matching
objective, verify live state and retrieve decisions with their corrections.
Preserve original wording, lineage and open/tentative status. A stale note or
compacted chat is evidence to reconcile; neither outranks the current request.
If only conversation context exists, save material context before it can be lost.
- State the settled scope, excluded actions and endpoint.
- Compose
promoteandto-docsfor supported claims into the existing authorized owners. Promotion may complete the request without implementation. - Compose
to-specfor required capability specifications andto-ticketsonly for their authorized delivery slices. Use the existing owner rather than restarting settled decisions or manufacturing work. - Verify and persist the achieved documents with
saveunder the authorized recovery boundary. A requested handoff composeshandoffas Markdown and carries the same scope. Stop here when this is the requested endpoint. - Only where implementation is authorized, compose
carryandimplementfor the selected tickets. Keep one lead responsible for evidence and docs; verify, obtain independent integration review and prove remote containment before claiming delivery. Continue independent authorized slices if another has a genuine blocker. Never select unrelated work. - Verify durable read-back before cleanup. Keep unresolved decisions, corrections and active handoff dependencies in the local JSON note.
Report the achieved endpoint, actual persistence results and any remaining gate. A rejected request, failed command or missing environment is not success and must not be converted into new authority by a helper.
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.
- today Changed · -51 lines · -63 tokens per session 4e34875105b3
- yesterday Changed · +8 lines 7a69f395a3f9
- 3d ago First seen · 83 lines · 101 tokens per session scan A e583a30e19c6
make-it-so is a skill published in the GitHub repository KaydenClark/LLM_Workbench (2 stars, last pushed today), licensed MIT. It adds 38 tokens to every session and 481 once invoked, about $0.0002 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 2026-09-07.
Other skills, from other repositories
gsd-quick
Execute a quick task with GSD guarantees (atomic commits, state tracking) but skip optional agents.
gsd-quick-batch
Batch several /gsd:quick-shaped tasks together — planned, dispatched, and merged as one run.
gh-issues
Fetch GitHub issues, spawn sub-agents to implement fixes and open PRs, then monitor and address PR review comments. Usage: /gh-issues [owner/repo] [--label bug] [--limit 5] [--milestone v1.0] [--assignee @me] [--fork user/repo] [--watch] [--interval 5] [--reviews-only] [--cron] [--dry-run] [--model glm-5]…
gsd-pause-work
Create context handoff when pausing work mid-phase.
gsd-ship
Create PR, run review, and prepare for merge after verification passes.
mcore-split-pr
Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.