matt-bat/got-skills / skills/help
help
VerifiedExplain the installed Agent Command Center skillpack when the user sends --help, including every active skill, activation marker, supported command, and concise behavior description. Also supplies the exact one-line startup hint required when the pack is first loaded into a conversation; do not activate for unrelated help requests.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.9 KB |
| agents/openai.yaml | config | 206 B |
| references/runtime-protocol.md | markdown | 3.1 KB |
| scripts/runtime_adapter.py | script | 14.3 KB |
references/runtime-protocol.md
View on GitHub# Host Runtime Protocol
Use this protocol when a host can execute local adapters. It makes startup and
mode behavior deterministic while leaving semantic task routing to the
canonical typed router. Explicit mode preferences persist across conversations
in a local JSON file; lifecycle startup signaling remains conversation-local.
## Event
Send one JSON object to `scripts/runtime_adapter.py` on stdin:
```json
{
"schema_version": 1,
"event": "user_message",
"message": "--clean-slate",
"state": null
}
```
`event` is either `pack_loaded` or `user_message`. Omit `message` for
`pack_loaded`. Omit `state` on the first event; afterwards, pass the exact
`state` object returned by the previous event. Durable preferences are reloaded
on every event, so toggles apply in later conversations. Set
`AGENT_COMMAND_CENTER_PREFERENCES` to a writable path for tests or a host
profile; otherwise the adapter uses `~/.config/agent-command-center/preferences.json`.
## Result
The adapter returns:
1. `startup_messages`: emit each string in order before other assistant output
2. `notices`: concise state changes or command errors to surface to the user
3. `recognized_command`: the owning skill or `null`
4. `consume_message`: whether the exact command is complete and should not be
treated as an ordinary task
5. `source_sha256`: binding to the exact user message
6. `descriptor_patch`: apply to the normalized task descriptor before routing
7. `descriptor_patch_policy`: requires a deep merge that unions, rather than
replaces, `constraints.explicit_skills`
8. `state`: pass unchanged into the next event for that conversation
Reject malformed adapter input. Do not silently repair or partially apply it.
Unknown commands and invalid forms of known commands are not consumed and do
not mutate state, so the complete message remains available to ordinary task
handling.
Use `apply_descriptor_patch()` when integrating in Python. Equivalent hosts must
preserve host-normalized explicit skills, append adapter-selected mode skills in
order without duplicates, and apply scalar action fields from the patch.
## Supported Commands
The executable grammar is:
```text
--help
--quizme [--mc] [--one-at-a-time] [--confirm] [--record]
--internal-lang on|off
--internal-lang --response on|off
--clean-slate
--clean-slate off
--ill-run-scripts
```
`--ill-run-scripts` is a bare toggle. When on, it pauses before easily executable scripts and gives the user an exact command and working directory.
All user-facing activation commands begin with `--`. Exact-command parsing does
not infer semantic task fields from keywords. The host must still normalize the
ordinary request into `task-descriptor.schema.json`, preserve the message digest
with its trace, and route every unresolved interpretation through
`eliminate-assumptions`.
## Conformance
Run:
```sh
python3 -m unittest skills.skill-governance.tests.test_runtime_adapter -v
```
A host that cannot execute this adapter must implement the same input, output,
state-transition, exact-command, and fail-closed behavior. Do not claim runtime
conformance from documentation alone.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |