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 CATWILLgh/MAINFRAME --skill mainframe-peer-workgit clone --depth 1 https://github.com/CATWILLgh/MAINFRAMEWrote 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/catwillgh/mainframe/mainframe-peer-work)<a href="https://agentmods.dev/skills/catwillgh/mainframe/mainframe-peer-work"><img src="https://agentmods.dev/badge/skills/catwillgh/mainframe/mainframe-peer-work/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/catwillgh/mainframe/mainframe-peer-work"><img src="https://agentmods.dev/badge/skills/catwillgh/mainframe/mainframe-peer-work.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 | $0.00024 | $0.00502 |
| Opus 5.5 | $0.00010 | $0.00201 |
| Sonnet 5.5 | $0.00005 | $0.00100 |
| Haiku 4.5 | $0.00002 | $0.00050 |
Grade A, and why
mainframe-peer-work 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 14d 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.
What it actually says
Work with an external agent peer
Use only the product selected by the user or by an applicable project agreement. Availability of a binary, login, subscription, or model is not authority to invoke it. Do not substitute a different product, install a CLI, create an account, change authentication, enable sharing, or broaden permissions implicitly.
Read only the selected product reference. Verify version-sensitive flags and permission behavior against installed CLI help; consult current official documentation for unresolved behavior. Reuse that evidence for subsequent calls to the same installed version:
Give the peer one bounded result, the exact project root, relevant evidence, permitted files and actions, prohibited external effects, verification requirements, and the recipient of its result. A review remains read-only. All task descriptions, questions, progress, evidence, corrections, and handoffs exchanged with the peer are in English.
Use the product's native non-interactive or headless interface and machine-readable output when available. Preserve the user's configured model and permission posture unless the user or project agreement selected another supported value. Never use a dangerous approval bypass by default.
Capture the exact session or conversation identifier from the first run. Resume that identifier for a correction or continuation of the same bounded result; do not use a most-recent shortcut when work may overlap. Start a new session for a different result or after acceptance. Use background execution only when the user or project agreement selected it and the product provides a reliable completion signal. Prefer native waiting, attach, or process completion over polling loops and adapter-specific helper state.
Separate concurrent writers by explicit file ownership or an authorized isolated checkout. Do not let two peers edit the same files opportunistically. Inspect the peer's actual output, changes, and checks before accepting its claims. Exit status, a success envelope, or a fluent summary alone is not proof of the assigned result.
If the selected CLI, login, permission boundary, session continuation, or completion signal is unavailable, return the exact missing operator action or capability. Continue directly only when doing so still satisfies the agreement and required independence.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 14d ago Changed · +3 lines · -33 tokens per session aced0eb02617
- 1mo ago First seen · 27 lines · 57 tokens per session scan A 6877b752e6d5
mainframe-peer-work is a skill published in the GitHub repository CATWILLgh/MAINFRAME (2 stars, last pushed 3d ago), licensed MIT. It adds 24 tokens to every session and 502 once invoked, about $0.0001 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 2026-09-08.
Other skills, from other repositories
minions
PM Ops — orchestrator of the minions AI subagent team. Single entry point: decomposes requests, delegates to minions- subagents in parallel or sequence, aggregates output.
optimize-goal
Use when the user wants to review, critique, tighten, or enhance an existing goal.md file (a written spec/prompt authored for a /goal command instead of typed into chat) before it gets executed — including requests like "review this goal.md", "enhance my goal.md", "optimize this goal", or sharing a goal.md file and…
ralplan
Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution.
remember
Review reusable project knowledge and decide what belongs in project memory, notepad, or durable docs.
atlas-obscura
Search Atlas Obscura for weird, wonderful, and hidden gem places near any destination. Find the interesting stuff, not boring plaques. Search by coordinates, get full details with descriptions and images.
cmux-architecture
Skill "cmux-architecture" from manaflow-ai/cmux, covering cmux architecture, package architecture, layers, file organization and documentation.