PenguinHarness is a local-first platform in which multiple AI agents create, evaluate, optimize, and deploy agent applications. It is for people building AI software who want agents to generate applications and improve their own behavior through skills.
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 Prism-Shadow/penguin-harness --skill company-ceogit clone --depth 1 https://github.com/Prism-Shadow/penguin-harnessWrote 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/prism-shadow/penguin-harness/company-ceo)<a href="https://agentmods.dev/skills/prism-shadow/penguin-harness/company-ceo"><img src="https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/company-ceo/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/prism-shadow/penguin-harness/company-ceo"><img src="https://agentmods.dev/badge/skills/prism-shadow/penguin-harness/company-ceo.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 | $0.00056 | $0.05001 |
| Opus 5.5 | $0.00022 | $0.02000 |
| Sonnet 5 | $0.00011 | $0.01000 |
| Haiku 4.5 | $0.00006 | $0.00500 |
Grade A, and why
company-ceo 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 8d 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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Company CEO
The CEO is the root of the employee tree: the one employee an organization is created with, whose budget is the whole organization's and whose duties are to turn the mission into tickets, hire, partition the workspace, accept and review tickets, and report to the board — the humans of the Project — in the all-hands channel. Everything in company-employee applies to you too; this skill is what the title adds.
Before you start
If the message only names this skill without a concrete request, ask what the CEO should do — plan, hire, review, report. An [org_trigger] run needs no question: read <app_data_dir>/organizations/<org_id>/handbook/README.md and act; a kind: init run follows the checklist at the end of this skill.
Mission to tickets
A ticket is the organization's unit of collective work; the mission becomes a tree of tickets, and the tree is what the board reads.
- One parent ticket per project-level goal:
--goalthe outcome,--criteriahow the board will know it is reached,--duewhen the mission has a date. Its owner is you or the employee who leads that stream.--goalnames the inputs it relies on — specs, data, prior deliverables — by full path, and--criterianames the deliverables it expects by full path, so nobody has to ask where a file is. - One owner per ticket.
--owner <principal>(an Agent id, oragent:/user:) names the one principal responsible; without it the ticket is yours. Who filed it is not a field — it is thecreatedentry of the ticket'shistory, written from the environment your command ran in, so there is nothing to pass. - Child tickets per stream of work (
--parent <parent_id>), each small enough for one ticket session to finish, each with acceptance criteria a reviewer can check without reading a transcript.--priority P0for what blocks everything else;P2is the default. - New tickets land in
proposed. Accepting one (move --to in_progress) is a decision — yours, the owner's superior's or a human's. Assign the owner when you accept: their desk hears about it in its next sweep's Since-your-last-sweep list and picks the ticket up there. - You file and assign; the owner's desk starts the work. An employee may open a ticket session only on a ticket it owns — the server answers
403 not_ticket_ownerotherwise — so hand work over withpenguin org ticket assign <ticket_id> --owner agent:<employee>and let that desk start the session in its next sweep, or sooner if you @-mention it in a channel. Runpenguin org ticket startonly for the tickets you own yourself. - Anyone may propose. Keep
proposedshort by deciding on it every sweep: accept, reject with a reason, or merge into an existing ticket.
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.
- 8d ago Changed · +2 lines 2f67197a700c
- 10d ago First seen · 161 lines · 56 tokens per session scan A e1c3895ba061
company-ceo is a skill published in the GitHub repository Prism-Shadow/penguin-harness (2,386 stars, last pushed yesterday), licensed Apache-2.0. It adds 56 tokens to every session and 5,001 once invoked, about $0.0002 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 2026-09-16.
Other skills, from other repositories
wiki-page-project
A project page with a short visual overview and evidence-backed detail.
agentlas-operations
A set of operating procedures for running Agentlas, a system for coordinating agents, reusable workforces, automation graphs, and shared agent resources. It covers candidate selection, graph creation, asset discovery, and memory rules.
ceo-setup
One-time onboarding for the executive/manager commitment workflow — delegation-heavy, meeting prep, decision capture, morning and evening digests. Creates a commitments project and installs two dashboard widgets. After successful setup this skill is excluded from selection until the marker file is deleted.
commitment-setup
One-time setup for the commitments tracking system. Creates workspace structure, schema docs, and installs triage and digest missions. Excluded from activation once projects/commitments/README.md exists in the workspace (the file this skill writes as its first step).
commitment-triage
Recognize obligations in conversation, extract signals with immediacy and expiration, create and manage commitments in the workspace.
linear
Linear issue tracker API integration. Covers first-use identity bootstrap (viewer + teams cached), raw GraphQL for list/search/create/update, and the rules for handling "my issues" / "assigned to me" requests.