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 rexleimo/aios --skill aios-harnessgit clone --depth 1 https://github.com/rexleimo/aiosWrote 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/rexleimo/aios/aios-harness)<a href="https://agentmods.dev/skills/rexleimo/aios/aios-harness"><img src="https://agentmods.dev/badge/skills/rexleimo/aios/aios-harness/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/rexleimo/aios/aios-harness"><img src="https://agentmods.dev/badge/skills/rexleimo/aios/aios-harness.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 | $0.02303 |
| Opus 5.5 | $0.00022 | $0.00921 |
| Sonnet 5.5 | $0.00011 | $0.00461 |
| Haiku 4.5 | $0.00006 | $0.00230 |
Grade A, and why
aios-harness 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 7d 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 — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AIOS Harness
Input and Output Contract
Use for a planned resumable objective selected by AIOS policy. Accept the current
work item, explicit workflow_type: one-shot|recurring, existing requirements/spec
refs, current Rex Command when present, and any selected recovery checkpoint.
The model declares task type; code must not infer it from prose or keywords.
Return a compact handoff with workflow_ref, workflow_revision, run_ref,
evidence_refs, recovery_checkpoint_ref, decision_brief, and
improvement_candidates. Use null or empty lists for inapplicable fields; do
not invent refs. The operating contract belongs in the existing spec/work item;
execution state belongs in the existing run journal. Reuse them rather than
creating another plan, wiki, or ledger.
Workflow Integration
Read references/workflow-integration.md when authoring or importing a recurring
workflow, preparing a human decision, or converting upstream loop-me/chief-of-staff
artifacts into AIOS handoffs.
- Specify: reuse accepted requirements. If a consequential decision is still
unresolved, report it to the current Rex Command; execute requirements work
only when Rex selects it. An explicit
/grillis the existing user entry. - Execute: once the operating contract is actionable, host only the current Command and persist its evidence. Do not turn the harness into a Provider chain.
- Decide: prepare authorized work before a human decision point. Return the evidence, options, recommendation, and exact pending decision in one brief.
- Learn: at a milestone, return source-backed improvement candidates and
write verified knowledge through existing Memo tools.
/retroremains the explicit environment-review entry; unselected candidates are not applied.
The contract records objective/non-goals, trigger/input, artifacts/completion evidence, failure/stop rules, budgets, and human decision points. For recurring work, distinguish the reusable definition and revision from each individual run. A previous run's success is not evidence for the next run. Definition changes are explicit revisions, not silent changes to an active run's acceptance.
What ships with it
4 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.
- 7d ago First seen · 157 lines · 56 tokens per session scan A ee4ba19c14be
aios-harness is a skill published in the GitHub repository rexleimo/aios (62 stars, last pushed yesterday), licensed MIT. It adds 56 tokens to every session and 2,303 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 2026-10-31.
aios-harness compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| aios-harnessrexleimo/aios | Skill | 62 | yesterday | MIT | A | 56 | Claude Code, Codex |
| live-narratorgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 34 | Claude Code, Codex |
| narrator-as-builtgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 29 | Claude Code, Codex |
| narrator-decisionsgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 26 | Claude Code, Codex |
| narrator-overviewgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 26 | Claude Code, Codex |
| narrator-phasesgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 27 | Claude Code, Codex |
| run-narratorgregce/deadreckon | Skill | 15 | 2mo ago | none | A | 52 | Claude Code |