Experiential is an open-source model gateway and router, meaning a service that gives agents one API for hosted, user-provided, local, and custom language models. It is for teams that need to choose models, control access and spending, and route production requests according to quality, speed, or cost. The catalogue entries provide agent skills and instructions for operating or configuring the gateway.
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 experientiallabs/experiential --skill test-buildgit clone --depth 1 https://github.com/experientiallabs/experientialWrote 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/experientiallabs/experiential/test-build)<a href="https://agentmods.dev/skills/experientiallabs/experiential/test-build"><img src="https://agentmods.dev/badge/skills/experientiallabs/experiential/test-build/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/experientiallabs/experiential/test-build"><img src="https://agentmods.dev/badge/skills/experientiallabs/experiential/test-build.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Data Exfiltration · line 308 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00076 | $0.04539 |
| Opus 5.5 | $0.00030 | $0.01816 |
| Sonnet 5.5 | $0.00015 | $0.00908 |
| Haiku 4.5 | $0.00008 | $0.00454 |
Grade C, and why
test-build scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "$WORK" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sSL --fail -H "Authorization: Bearer $HF_TOKEN" \ How it starts
The opening of the file, as written. The whole thing — 357 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Test the build path
Execute the current CLI path on one isolated local root. This is a live provider run, not the deterministic fixture test. It proves the commands complete, that the fitted router replays exactly, and that routed loopback traffic journals durably. It does not claim router quality.
Every step is an exp or standard shell command. Do not edit models.toml, project files, or
artifacts by hand. Do not call internal Python APIs. If a step cannot be done through the CLI,
stop and report the UX gap.
Do not use OpenRouter. Do not run exp optimize model. Do not start E2B sandboxes. Do not commit
.exp/, traces, or any generated artifact.
Time every numbered step and report the table at the end.
Fit dataset
This skill fits the pinned public terminal-tasks export by default. A caller may override any variable below; when overriding, the caller owns the digest. Record the effective values in the final report so the fitted dataset is unambiguous.
EXP_TEST_BUILD_PROJECT(defaultterminal-tasks): local project ID passed to every commandEXP_TEST_BUILD_TRACES(default$HOME/exp-fixtures/terminal-tasks-traces.otel.jsonl): local trace-export fileEXP_TEST_BUILD_TRACES_SHA256(default21c62cba7e3372cbf03df051dc2408699fbf8ea3561ba661b599e4949f0e5d42): lowercase hex digestEXP_TEST_BUILD_SOURCE(defaultotlp): trace source format
When the default file is absent, download the pinned export once with HF_TOKEN:
https://huggingface.co/datasets/experiential-labs/wmo-terminal-tasks-traces/resolve/540883e451dc13d34fb50fdd36b143cb0f1fb0db/traces.otel.jsonl
Do not skip the digest check. If the file is absent and cannot be downloaded, or the digest does not match, stop.
Optional work root: EXP_TEST_BUILD_ROOT (default /tmp/exp-test-build). Keep the trace file
outside that directory. The skill deletes the work root at the start of the run.
Secrets
Fail before any write if a required variable is missing or empty. Never print a secret value.
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.
- 1mo ago First seen · 357 lines · 76 tokens per session scan C a53bd7c269cd
test-build is a skill published in the GitHub repository experientiallabs/experiential (9,911 stars, last pushed 11d ago), licensed Apache-2.0. It adds 76 tokens to every session and 4,539 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.