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/sb_conflicts.py
View on GitHub"""Cheap contradiction check run at `promote` time.
Two facts in the same scope conflict when they talk about the same thing
(Jaccard overlap of normalised content words >= CONFLICT_JACCARD) with
opposite polarity (exactly one of them contains a negation such as
never / don't / do not / no / avoid / not). Deliberately lexical and
conservative: antonym pairs (tabs vs spaces) are out of scope.
"""
from __future__ import annotations
import re
CONFLICT_JACCARD = 0.4
NEGATIONS = frozenset(
"never not no avoid don't dont doesn't doesnt do-not mustn't mustnt shouldn't "
"shouldnt can't cant cannot won't wont without forbid forbidden disallow".split()
)
STOPWORDS = frozenset(
"a an the and or but of to for in on at by with from into onto as is are was "
"were be been it its this that these those there here than then so if when "
"always must should use using used do does please only just all any each "
"you we they i your our via per e g eg ie".split()
)
_TOKEN = re.compile(r"[a-z0-9]+(?:'[a-z]+)?")
def _tokens(text: str) -> list[str]:
return _TOKEN.findall(text.lower().replace("’", "'"))
def negated(text: str) -> bool:
return any(t in NEGATIONS for t in _tokens(text))
def content_words(text: str) -> set[str]:
out = set()
for t in _tokens(text):
if t in NEGATIONS or t in STOPWORDS or len(t) < 2:
continue
if len(t) > 3 and t.endswith("s") and not t.endswith("ss"):
t = t[:-1] # crude plural folding: builds == build
out.add(t)
return out
def jaccard(a: set[str], b: set[str]) -> float:
if not a or not b:
return 0.0
return len(a & b) / len(a | b)
def conflicts(text: str, scope: str, facts: dict, skip_keys: set[str] = frozenset(),
threshold: float = CONFLICT_JACCARD) -> list[tuple[str, float]]:
"""(fact_id, overlap) for every fact in `scope` that `text` contradicts."""
words, neg = content_words(text), negated(text)
out = []
for fid, f in facts["facts"].items():
if f.get("scope") != scope or f.get("key") in skip_keys:
continue
if negated(f["representation"]) == neg:
continue
overlap = jaccard(words, content_words(f["representation"]))
if overlap >= threshold:
out.append((fid, overlap))
out.sort(key=lambda x: -x[1])
return out
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 |