moonlight-lupin/agent-skills / agent-ops/operator-brain
operator-brain
UnverifiedUse when communicating, deciding, or reporting as a senior operator — 10 behavioral modules from Anthropic's Fable 5 and Fable 5.1 prompting guides. Not capability — pure behavior. Load for user-facing text quality, narrating long work, and keeping edits in scope.
| File | Kind | Size |
|---|---|---|
| EVALUATION.md | markdown | 5.8 KB |
| SKILL.md | skill md | 10.2 KB |
| evals/fixtures.py | script | 1.7 KB |
| evals/harness.py | script | 9.8 KB |
| evals/tasks/T1_research.txt | markdown | 391 B |
| evals/tasks/T2_files.txt | markdown | 224 B |
| evals/tasks/T3_missing.txt | markdown | 104 B |
| evals/tasks/T4_edit.txt | markdown | 190 B |
evals/fixtures.py
View on GitHub"""Workspace fixtures for the operator-brain eval battery. Same files every run."""
import os
def seed(ws):
"""Seed the four-file workspace: three notes + one config helper.
Refuses to seed into a non-empty directory unless FIXTURES_ALLOW_OVERWRITE=1 (seeding overwrites the four fixture filenames)."""
os.makedirs(ws, exist_ok=True)
if os.listdir(ws) and os.environ.get("FIXTURES_ALLOW_OVERWRITE") != "1":
raise RuntimeError(f"refusing to seed into non-empty workspace {ws}; set FIXTURES_ALLOW_OVERWRITE=1 to override")
open(f"{ws}/notes_a.txt", "w").write(
"deploy plan v2\nTODO: confirm rollback window\nstep 1 build image\nstep 2 push tag\n"
"step 3 migrate db\nTODO: page oncall before step 3\nstep 4 smoke test\nstep 5 announce\n"
"step 6 retro\nschedule: fri 21:00\nowner: infra\nappendix: runbook link\n")
open(f"{ws}/notes_b.txt", "w").write(
"standup notes\nalice: api latency down\ncarol: flaky test fixed\n"
"next: load test thursday\nsnacks: someone owes cake\n")
open(f"{ws}/notes_c.txt", "w").write(
"incident 42 review\nroot cause: cache stampede\nTODO: add jitter\nTODO: alert on p99\n"
"follow up in sprint 12\nstatus: mitigated\nlabels: cache, outage\nimpact: 14 min\n")
open(f"{ws}/config.py", "w").write(
'''import os
import json # TODO: unused import?
# TODO: move to env before production
API_KEY = "sk-live-hardcoded-9f2c41"
DEFAULTS = {
"cache_ttl": 300,
"max_retries": 3,
"log_level": "info",
}
def get_setting(name, overrides=None):
cfg = dict(DEFAULTS)
if "cache_ttl" not in cfg:
raise KeyError("cache_ttl missing")
if overrides:
cfg.update(overrides)
return cfg.get(name)
''')
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: 472k | 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 |