Harmix/open-stanley / open-stanley/skills/stanley
stanley
VerifiedYour head of content. Use this whenever the user wants help with LinkedIn or X (Twitter) — writing or scheduling a post, finding things to comment on, turning a meeting, voice note, or Slack thread into a post, reviewing why a post did or didn't work, planning the week's content, or asking "what should I post". Also use it for anything that mentions Stanley, ghostwriting, personal brand, founder-led content, building in public, or growing followers, even if the user doesn't say "post". Start here; it routes to the other open-stanley skills and holds the operating rules they all share.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 14.2 KB |
| references/capabilities.md | markdown | 3.3 KB |
| references/platform-access.md | markdown | 4.0 KB |
| references/recommended-mcps.md | markdown | 5.5 KB |
| references/recommended.mcp.json | config | 2.0 KB |
| references/research-linkedin-2026.md | markdown | 6.7 KB |
| references/scheduling.md | markdown | 8.6 KB |
| references/vault-backends.md | markdown | 6.8 KB |
| references/verify-live.md | markdown | 2.2 KB |
| references/why-these-rules.md | markdown | 2.1 KB |
| scripts/stanley-auditexecutable | script | 24.6 KB |
| scripts/stanley-briefexecutable | script | 6.4 KB |
| scripts/stanley-chartexecutable | script | 14.5 KB |
| scripts/stanley-previewexecutable | script | 5.8 KB |
| scripts/stanley-statsexecutable | script | 14.9 KB |
| scripts/stanley-trace | other | 6.3 KB |
| scripts/stanley-vaultexecutable | script | 35.1 KB |
scripts/stanley-brief
View on GitHub#!/usr/bin/env python3
"""stanley-brief: the human-pass contract.
Two commands.
make STORY.md --claim "one sentence" [--vault DIR] [--platform linkedin|x] [--length 180]
→ prints a JSON brief for a human writer: the claim, the facts with sources, required
and forbidden terms, length, ending style, and a short negative style guide drawn
from voice.md. No prose. A human who is handed prose lightly edits it; a human who
is handed a brief writes, and only written text is free of any model watermark.
verify BRIEF.json HUMAN.txt [--draft CLAUDE_DRAFT.txt]
→ checks the returned text: every fact string present, no forbidden term, length in
band, ending style honored, and (if --draft given) low 3-gram overlap with the model
draft so a "rewrite" that is really a copy is caught. Exit 1 on failure.
Facts are matched literally (numbers and quoted phrases), so keep them short in the story file.
"""
import argparse, json, re, sys
from pathlib import Path
def frontmatter(text):
m = re.match(r"^---\n(.*?)\n---\n(.*)$", text, re.S)
meta = {}
body = text
if m:
for line in m.group(1).splitlines():
if ":" in line:
k, v = line.split(":", 1); meta[k.strip()] = v.split("#", 1)[0].strip()
body = m.group(2)
return meta, body
def voice_rules(vault):
rules = ["Open on the event or the claim; no setup sentence.", "No 'not X, it's Y'.", "No list of three.",
"No summary or 'in short' ending.", "No hashtags. No links in the body.", "Vary sentence length; no three short sentences in a row."]
forbidden = ["delve", "leverage", "robust", "pivotal", "seamless", "game-changer", "here's the thing", "let that sink in",
"in today's", "moreover", "furthermore", "crucial", "unlock", "elevate", "navigate", "landscape", "testament"]
ending = "flat"
if vault:
p = Path(vault) / "voice.md"
if p.exists():
t = p.read_text(encoding="utf-8")
y = re.search(r"```yaml\n(.*?)```", t, re.S)
if y:
for line in y.group(1).splitlines():
if line.startswith("banned_extra"):
forbidden += [x.strip() for x in line.split(":", 1)[1].split("#")[0].strip("[] ").split(",") if x.strip()]
if line.startswith("em_dash") and "ban" in line: rules.append("No em dashes.")
if line.startswith("closing_question") and "allow" in line: ending = "flat or one specific open question"
how = re.search(r"## How this person writes.*?\n(.*?)\n## ", t, re.S)
if how:
rules += [l.strip("- ").strip() for l in how.group(1).splitlines() if l.strip().startswith("-")][:6]
return rules, sorted(set(forbidden)), ending
def facts_from(body):
facts = []
for m in re.finditer(r"\b\d[\d,.]*\s?(?:%|out of \d+|/\d+|[kKM]\b|x\b)?", body):
s = m.group(0).strip()
if len(s) > 1 and s not in facts: facts.append(s)
for q in re.findall(r"[\"“]([^\"”]{8,140})[\"”]", body): facts.append(q.strip())
return facts[:12]
def cmd_make(a):
meta, body = frontmatter(Path(a.story).read_text(encoding="utf-8"))
rules, forbidden, ending = voice_rules(a.vault)
brief = {
"task": "Write this post yourself, in your own words, from the material below. Do not paste or lightly edit any text you were given elsewhere; every sentence should be yours.",
"platform": a.platform,
"claim": a.claim,
"material": body.strip()[:2500],
"facts_to_keep_exactly": facts_from(body),
"source": meta.get("source", ""),
"sensitivity": meta.get("sensitivity", "anonymize"),
"required_terms": [t for t in (a.require or [])],
"forbidden_terms": forbidden,
"length_words": [int(a.length * 0.75), int(a.length * 1.25)],
"ending": ending,
"style_rules": rules,
"first_line": "Must be a complete thought that works on its own; it is all a reader sees before clicking 'more'.",
}
print(json.dumps(brief, indent=1, ensure_ascii=False))
def shingles(t, n=3):
w = re.findall(r"[a-z0-9']+", t.lower()); return {" ".join(w[i:i+n]) for i in range(max(0, len(w)-n+1))}
def cmd_verify(a):
brief = json.loads(Path(a.brief).read_text(encoding="utf-8"))
text = Path(a.human).read_text(encoding="utf-8").strip()
problems, checks = [], []
low = text.lower()
for f in brief.get("facts_to_keep_exactly", []):
if f.lower() not in low: checks.append(f"fact not found verbatim (dropped on purpose, or changed?): {f!r}")
for t in brief.get("required_terms", []):
if t.lower() not in low: problems.append(f"required term missing: {t!r}")
for t in brief.get("forbidden_terms", []):
if re.search(r"\b" + re.escape(t.lower()) + r"\b", low): problems.append(f"forbidden term used: {t!r}")
n = len(text.split()); lo, hi = brief.get("length_words", [0, 10**9])
if not (lo <= n <= hi): problems.append(f"length {n} words outside {lo}-{hi}")
if brief.get("ending") == "flat" and text.rstrip().endswith("?"): problems.append("ends with a question but brief asked for a flat ending")
if re.search(r"(?<!\w)#\w+", text): problems.append("hashtag present")
if a.draft:
d = Path(a.draft).read_text(encoding="utf-8")
s1, s2 = shingles(text), shingles(d)
j = len(s1 & s2) / max(1, len(s1 | s2))
if j > 0.25: problems.append(f"3-gram overlap with the model draft is {j:.0%}; this is an edit, not a rewrite")
else: print(f"overlap with model draft: {j:.0%} (good, this is new text)")
for c in checks: print("CHECK", c)
for p in problems: print("FAIL", p)
print("RESULT:", "PASS" if not problems else f"{len(problems)} problem(s)")
sys.exit(1 if problems else 0)
def main():
ap = argparse.ArgumentParser(); sp = ap.add_subparsers(dest="cmd", required=True)
p = sp.add_parser("make"); p.add_argument("story"); p.add_argument("--claim", required=True); p.add_argument("--vault")
p.add_argument("--platform", default="linkedin"); p.add_argument("--length", type=int, default=180); p.add_argument("--require", action="append"); p.set_defaults(f=cmd_make)
p = sp.add_parser("verify"); p.add_argument("brief"); p.add_argument("human"); p.add_argument("--draft"); p.set_defaults(f=cmd_verify)
a = ap.parse_args(); a.f(a)
if __name__ == "__main__":
main()
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 |