kernel-goal-loop SKILL.md skillA
kernel-goal-loop from tedix-hq/tedix is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add tedix-hq/tedix/kernel-goal-loop@git:20261006.8886800Pins git:20261006.8886800 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- tedix-hq/tedix
- Path
apps/skill-runtime/examples/kernel-goal-loop/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261006.8886800, committed Oct 6, 2026
- Size
- 3,019 bytes
- Repository stars
- 27
- License
- AGPL-3.0
Durable, resumable reference L2 goal-loop for deterministic or adversarial text evaluation, persisting every turn as a durable Workflow step. Objective Work Item evaluation belongs to the first-class kernel goal-loop Workflow.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# kernel-goal-loop
The durable form of the **L2 goal-loop** primitive: a maker turn repeats until a
separate evaluator says the goal is met or a bound is hit. Where the `kernel_goal_loop` Code Mode
muscle polls inside a single ~30s isolate, this skill runs the same governed
loop as a **Cloudflare Workflow**: each kernel turn, poll, and inter-turn wait is
a durable `step.do` / `step.sleep`, so the loop survives DO hibernation, retries
per step, and is observable through the run timeline + artifacts.
## What it does
Each iteration (up to `maxTurns`):
1. `step.do` — submit a kernel turn via `home.ask({ content })` (the
maker — makes progress on the goal every turn, regardless of evaluator).
2. `step.do` + `step.sleep` — poll `home.read_home_run` until the run is
terminal, extracting only the small route/cost fields (the large run-set
payload never lands in a persisted step value).
3. **Separate-evaluator stop-check** (checker ≠ maker):
- `deterministic` — a regex the maker prompt never sees, tested against
the maker's answer.
- `adversarial` — a second `ask` judge turn that may only rule
DONE/CONTINUE and is forbidden from answering the task.
…Read the whole file at its exact version.
How to install
mdr add tedix-hq/tedix/kernel-goal-loop@sha256:2f3c5ccfe853d634The line at the top of this page pins the label git:20261006.8886800, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_aejnkosvygdf5yn6)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 6, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (3019 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
tedix-hq/tedix · 27 stars · license AGPL-3.0 · pushed 2026-10-25 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_aejnkosvygdf5yn6 GET https://markdownregistry.com/api/v1/resolve?ref=tedix-hq/tedix/kernel-goal-loop GET https://markdownregistry.com/api/v1/blob/2f3c5ccfe853d634d88343306ed0b4186e447694350e4e24ddebaef9435d49de
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.