oliverweissl/context-garden / skills/seedbank
seedbank
VerifiedMaintain a small, generated AGENTS.md of repository facts (build commands, conventions, invariants, past mistakes) that were rediscovered often or were costly to get wrong. Use at task start to check cached facts, and at task end to log and promote what you had to dig for.
| File | Kind | Size |
|---|---|---|
| .gitignore | other | 40 B |
| SKILL.md | skill md | 2.7 KB |
| agents/gardener.md | markdown | 1.6 KB |
| bin/seedbankexecutable | script | 1.2 KB |
| references/guide.md | markdown | 3.3 KB |
| references/schema.md | markdown | 5.4 KB |
| references/scoring.md | markdown | 2.1 KB |
| scripts/cg_fingerprint.py | script | 3.1 KB |
| scripts/cg_mode.py | script | 5.4 KB |
| scripts/sb_compile.py | script | 5.6 KB |
| scripts/sb_conflicts.py | script | 2.3 KB |
| scripts/sb_hook.py | script | 13.5 KB |
| scripts/sb_scoring.py | script | 3.4 KB |
| scripts/sb_store.py | script | 9.3 KB |
| scripts/seedbank.pyexecutable | script | 43.0 KB |
| tests/fixtures/mini_repo/CMakeLists.txt | markdown | 94 B |
| tests/fixtures/mini_repo/README.md | markdown | 126 B |
| tests/fixtures/mini_repo/src/generated/bindings.cpp | other | 92 B |
| tests/fixtures/mini_repo/tools/codegen.py | script | 57 B |
| tests/smoke_test.shexecutable | script | 29.7 KB |
scripts/cg_mode.py
View on GitHub#!/usr/bin/env python3
"""Per-skill mode: on (Claude uses it automatically), manual (only when the
user calls it, e.g. `/compost <instruction>`), off (not used; the default).
Identical copy in every context-garden skill (skills never import each other).
The mode is stored where Claude Code's own `skillOverrides` setting lives, so
the `/skills` menu and this tool agree:
on -> "on", manual -> "user-invocable-only", off -> "off" ("name-only" reads as on).
Resolution: $CONTEXT_GARDEN_MODE_<NAME> > <git root>/.claude/settings.local.json
> <git root>/.claude/settings.json > ~/.claude/settings.json > off.
Claude Code applies skillOverrides natively only to standalone skills (not
plugin skills), so SKILL.md also injects `mode --banner` and hooks check
`--is-on`; that makes the mode hold for plugin installs too.
Usage: cg_mode.py <skill> [on|manual|off] [--scope local|project|user]
cg_mode.py <skill> --banner | --is-on (exit 0 on, 3 manual/off)
"""
from __future__ import annotations
import json
import os
import shlex
import subprocess
import sys
import tempfile
from pathlib import Path
MODES = ("on", "manual", "off")
DEFAULT_MODE = "off"
NOT_ON = 3 # --is-on exit status when manual/off
BIN_DIR = Path(__file__).resolve().parent.parent / "bin"
TO_OVERRIDE = {"on": "on", "manual": "user-invocable-only", "off": "off"}
FROM_OVERRIDE = {"on": "on", "name-only": "on", "user-invocable-only": "manual", "off": "off"}
def repo_root() -> Path:
try:
out = subprocess.run(
["git", "rev-parse", "--show-toplevel"], stdin=subprocess.DEVNULL,
stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, universal_newlines=True,
)
if out.returncode == 0 and out.stdout.strip():
return Path(out.stdout.strip())
except OSError:
pass
return Path.cwd()
def settings_files() -> dict:
root = repo_root()
return {
"local": root / ".claude" / "settings.local.json",
"project": root / ".claude" / "settings.json",
"user": Path.home() / ".claude" / "settings.json",
}
def _read(path: Path) -> dict:
try:
data = json.loads(path.read_text())
except FileNotFoundError:
return {}
return data if isinstance(data, dict) else {}
def resolve(skill: str) -> tuple:
"""(mode, source)."""
env = os.environ.get(f"CONTEXT_GARDEN_MODE_{skill.upper()}")
if env in MODES:
return env, f"$CONTEXT_GARDEN_MODE_{skill.upper()}"
for scope, path in settings_files().items():
try:
value = (_read(path).get("skillOverrides") or {}).get(skill)
except ValueError:
continue # unparseable settings file: Claude Code will complain itself
if value in FROM_OVERRIDE:
return FROM_OVERRIDE[value], str(path)
return DEFAULT_MODE, "default"
def set_mode(skill: str, mode: str, scope: str = "local") -> Path:
"""Write skillOverrides[skill]; every other key of the file is kept as is."""
path = settings_files()[scope]
try:
data = _read(path)
except ValueError:
raise SystemExit(f"{path} is not valid JSON; not touching it")
overrides = data.setdefault("skillOverrides", {})
overrides[skill] = TO_OVERRIDE[mode]
path.parent.mkdir(parents=True, exist_ok=True)
fd, tmp = tempfile.mkstemp(dir=path.parent, prefix=f".{path.name}.", suffix=".tmp")
try:
with os.fdopen(fd, "w") as f:
f.write(json.dumps(data, indent=2) + "\n")
os.replace(tmp, path)
except BaseException:
try:
os.unlink(tmp)
except OSError:
pass
raise
return path
def banner(skill: str) -> str:
mode, _ = resolve(skill)
if mode == "manual":
return (
f"**{skill} is in MANUAL mode here.** Use it only if the user explicitly asked "
f"for it in this conversation (e.g. typed `/{skill} ...`). Otherwise stop reading "
"this skill and continue the task without it."
)
if mode == "off":
return (
f"**{skill} is turned OFF here** (off is the default). Do not use it; continue the "
f"task without it. Only if the user asks to turn it on: `{shlex.quote(str(BIN_DIR / skill))} mode on` "
"(automatic) or `mode manual` (only when called)."
)
return ""
def main(argv: list) -> int:
if not argv:
print(__doc__.strip())
return 2
skill, rest = argv[0], argv[1:]
if rest == ["--banner"]:
text = banner(skill)
if text:
print(text)
return 0
if rest == ["--is-on"]:
return 0 if resolve(skill)[0] == "on" else NOT_ON # a crash exits 1: told apart
if not rest:
mode, source = resolve(skill)
print(f"{skill}: {mode} (from {source})")
return 0
mode, scope = rest[0], "local"
if len(rest) == 3 and rest[1] == "--scope" and rest[2] in ("local", "project", "user"):
scope = rest[2]
elif len(rest) != 1:
print(__doc__.strip(), file=sys.stderr)
return 2
if mode not in MODES:
print(f"mode must be one of {', '.join(MODES)}", file=sys.stderr)
return 2
path = set_mode(skill, mode, scope)
print(f"{skill}: {mode} (written to {path})")
now, source = resolve(skill)
if now != mode:
print(f"note: still {now} because {source} takes precedence", file=sys.stderr)
return 0
if __name__ == "__main__":
sys.exit(main(sys.argv[1:]))
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 |