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 xingchengxu/OpenExpertise --skill skill-experience-creatorgit clone --depth 1 https://github.com/xingchengxu/OpenExpertiseWrote 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/xingchengxu/openexpertise/skill-experience-creator)<a href="https://agentmods.dev/skills/xingchengxu/openexpertise/skill-experience-creator"><img src="https://agentmods.dev/badge/skills/xingchengxu/openexpertise/skill-experience-creator/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/xingchengxu/openexpertise/skill-experience-creator"><img src="https://agentmods.dev/badge/skills/xingchengxu/openexpertise/skill-experience-creator.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 | $0.00166 | $0.01712 |
| Opus 5 | $0.00083 | $0.00856 |
| Sonnet 5 | $0.00033 | $0.00342 |
| Haiku 4.5 | $0.00017 | $0.00171 |
Grade A, and why
experience-creator 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 10d 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 — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Experience Creator
Turn a user goal into a runnable experience.yaml plus its accompanying
prompt files, tool modules, and dataset descriptors. The result is a directory
that oe run can execute end-to-end.
The deep material lives in three reference files — read them when the step says so:
references/api-reference.md— every field ofexperience.yaml, every node kind's options, every control-flow construct.references/patterns.md— copy-paste topologies (linear, fan-out, pipeline, bounded loop, dataset → agent → score).references/lessons.md— gotchas from earlier V1 work (schema enforcement, cache invalidation,exactOptionalPropertyTypes, ESM imports, etc.).
Starter templates are in assets/templates/. Two worked examples are in
assets/examples/. A standalone validator is scripts/validate-experience.mjs.
Step 0 — Confirm OpenExpertise is installed
oe --version
If oe is not on PATH, point the user at the OpenExpertise repo and tell them
to either install the CLI (npm install -g @openexpertise/cli) or run via the
workspace (node packages/cli/dist/bin.js).
Step 1 — Establish the goal (do NOT skip)
Before writing a line of YAML, get crisp answers to:
- What does this experience produce? A reviewed PR? A summarized research finding? A risk score? Name the artifact concretely.
- What's the unit of work? The thing one node does once: review one file, research one topic, score one transaction.
- What state does it need to remember across runs? This is the
blackboard's reason to exist (
past_incidents,tuned_thresholds,accumulated_findings). - Which nodes are LLM-backed vs deterministic? Map each step to one of the five kinds (agent / skill / tool / dataset / experience).
- What's "done"? A specific exit state — a final field set, a status reaching a value.
Write these answers down for the user before drafting.
Step 2 — Pick the topology
What ships with it
28 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.
- assets/examples/hello-tool/experience.yaml 464 B
- assets/examples/hello-tool/package.json 111 B
- assets/examples/hello-tool/README.md 222 B
- assets/examples/hello-tool/tools/greet.mjs 108 B runs code
- assets/examples/hello-tool/tools/greet.ts 126 B runs code
- assets/examples/README.md 731 B
- assets/examples/review-branch/experience.yaml 2.0 KB
- assets/examples/review-branch/package.json 114 B
- assets/examples/review-branch/prompts/review.md 124 B
- assets/examples/review-branch/prompts/score.md 85 B
- assets/examples/review-branch/prompts/verify.md 111 B
- assets/examples/review-branch/README.md 314 B
- assets/examples/review-branch/tools/list_dimensions.mjs 269 B runs code
- assets/templates/agent-with-schema.yaml 465 B
- assets/templates/dataset-plus-tool.yaml 460 B
- assets/templates/fan-out.yaml 649 B
- assets/templates/pipeline.yaml 800 B
- assets/templates/simple-tool.yaml 234 B
- commands/ultraexpertise.md 2.2 KB
- package.json 1.0 KB
- README.md 2.0 KB
- references/api-reference.md 4.6 KB
- references/lessons.md 2.5 KB
- references/patterns.md 2.7 KB
- scripts/validate-experience.mjs 1.0 KB runs code
- tests/skill-md.test.ts 769 B runs code
- tests/templates.test.ts 861 B runs code
- tests/validate-script.test.ts 768 B runs code
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.
- 10d ago First seen · 201 lines · 166 tokens per session scan A a056b4800373
experience-creator is a skill published in the GitHub repository xingchengxu/OpenExpertise (4 stars, last pushed 3mo ago), licensed MIT. It adds 166 tokens to every session and 1,712 once invoked, about $0.0008 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
hatch3r-board-groom
Ongoing backlog refinement for existing board items. Re-prioritize, reclassify, re-scope, archive stale items, decompose oversized issues, merge duplicates, refresh dependencies, and remediate board health findings.
hatch3r-cli-toolbox
Category-indexed reference for 34 specialist CLI tools beyond the always-on five (ripgrep, jq, gh, fd, fzf). Use to pick the right tool for HTTP clients, ai-chat, structural-search, sed-style edits, data ops, browser automation, container ops, and more.
hatch3r-board-init
Initializes a project board (GitHub Projects V2, Azure Boards, or GitLab Issue Boards) with hatch3r's label taxonomy, status fields, and board structure. Platform detected from hatch.json.
hatch3r-board-shared
Shared context and procedures for all board commands. Provides platform-agnostic board config, label taxonomy, branch conventions, sync enforcement, and tooling directives. Platform-specific details are in commands/board/shared-{platform}.md.
hatch3r-browser-verify
Opt-in browser verification skill — spec-run-first Playwright verification (assertions execute in the runner, agent reads only failures), axe-core a11y audits, toHaveScreenshot() regression diffs, E2E test scaffolds, and snapshot-mode exploratory driving. Default ON for UI-affecting agent invocations; disable globally…
hatch3r-board-refresh
Regenerates the living board overview dashboard from current board state. Scans all open issues, computes health metrics, and updates the meta:board-overview issue.