JPeetz/agent-skills / skills/skill-development/skill-lifecycle-foundry
skill-lifecycle-foundry
VerifiedUse this skill when mining, authoring, personalizing, or generalizing agent skills.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 595 B |
| LICENSE | other | 1.1 KB |
| SKILL.md | skill md | 7.6 KB |
| evals/evals.json | config | 4.0 KB |
| references/release-checklist.md | markdown | 1.2 KB |
| references/sources.md | markdown | 464 B |
| scripts/scan_sessions.py | script | 2.6 KB |
scripts/scan_sessions.py
View on GitHub#!/usr/bin/env python3
"""Scan a directory of session/archive files for repeated workflow intents.
A deterministic first-pass miner: walk the given root, read each text file
(.md, .txt, .json, .jsonl, .log, .yaml, .yml), and report how many distinct
files contain each workflow-signal keyword. Use it to shortlist candidate
workflows before drafting a SKILL.md.
Stdlib only. Invoke through the Hermes terminal tool:
python3 scripts/scan_sessions.py <root_dir> [--min-files 3] [--top 10]
Example:
python3 scripts/scan_sessions.py ~/.hermes --top 8
"""
from __future__ import annotations
import argparse
import collections
import pathlib
import re
DEFAULT_KEYWORDS = [
"review",
"migrate",
"refactor",
"summary",
"report",
"build",
"deploy",
"triage",
"optimize",
"backup",
]
TEXT_EXT = {".md", ".txt", ".json", ".jsonl", ".log", ".yaml", ".yml"}
def tokens(text: str) -> set[str]:
"""Return the set of lowercase alphanumeric tokens in a text blob."""
return set(re.findall(r"[a-z0-9_]+", text.lower()))
def scan(root: pathlib.Path, min_files: int, top: int) -> None:
if not root.is_dir():
raise SystemExit(f"not a directory: {root}")
hits: dict[str, set] = collections.defaultdict(set)
file_count = 0
for path in root.rglob("*"):
if not path.is_file() or path.suffix.lower() not in TEXT_EXT:
continue
file_count += 1
try:
text = path.read_text(encoding="utf-8", errors="ignore")
except OSError:
continue
found = tokens(text) & set(DEFAULT_KEYWORDS)
for kw in found:
hits[kw].add(str(path))
for kw in DEFAULT_KEYWORDS:
hits.setdefault(kw, set()) # stable output even when a word is unseen
print(f"scanned {file_count} files under {root}")
print(f"{'keyword':<14}{'files':>7} intent")
ranked = sorted(hits.items(), key=lambda kv: len(kv[1]), reverse=True)
for kw, files in ranked[:top]:
marker = " *candidate*" if len(files) >= min_files else ""
print(f"{kw:<14}{len(files):>7}{marker}")
def main() -> None:
parser = argparse.ArgumentParser(description="session intent scanner")
parser.add_argument("path", type=pathlib.Path, help="directory to scan")
parser.add_argument("--min-files", type=int, default=3,
help="distinct-file threshold to call a candidate")
parser.add_argument("--top", type=int, default=10, help="rows to print")
args = parser.parse_args()
scan(args.path, args.min_files, args.top)
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 |