Generate and keep project-specific agent configs synchronized as your codebase changes.
Project overview
Caliber analyzes a repository and creates tailored context files, rules, skills, and MCP settings for tools such as Claude Code, Cursor, Codex, OpenCode, and GitHub Copilot. After setup, refresh hooks keep those files aligned with code changes. Proposed edits are shown as diffs and backed up, while generation uses an existing agent subscription or a user-provided model key.
General skillSoftware developmentModel Context Protocol
kimiflow
@kimikonapps·Python
Enforce gated feature and bug-fix workflows in Claude Code or Codex.
Project overview
kimiflow runs an explicitly invoked feature or bug-fix loop through clarification, diagnosis, planning, implementation, verification, review, and commit approval. Tested scripts enforce blocker resolution, red-and-green evidence, durable state, and other gates instead of relying on the model’s self-report. It can also maintain a local codebase map and learning memory, with optional Obsidian integration; jq is required, and the workflow stops for explicit approval before committing.
Enforce maintainability gates around AI-assisted coding workflows.
Project overview
Slop-Mop adds an opinionated quality loop to repositories built with coding agents: run fast checks while working, a broader sweep before a pull request, and follow-up after CI or review feedback. Its gates target bogus tests, bypassed checks, coverage and typing gaps, complexity, duplication, security, and dependency risks, with temporary baselines for inherited debt.
Scans staged changes for secrets and teaches Claude Code to avoid hardcoding credentials.
Project overview
secret-hygiene teaches Claude Code to read credentials from the environment instead of hardcoding them, then uses a zero-dependency Python scanner on directories or staged changes. It detects JWTs, PEM keys, common cloud credentials, connection strings, and high-entropy secret-like variables, with narrowly scoped ignore rules. For public or production repositories, it recommends adding gitleaks and trufflehog.
Carry Claude Code project context across sessions with two Markdown files
Project overview
session-continuity keeps a project's memory in two committed Markdown files: a current-state SESSION_PRIMER and an append-only LEARNINGS log. Slash commands refresh the primer, record hard-won lessons, and close a session, while hooks surface stale context or missing checks. The approach stays portable and human-readable instead of relying on a hosted memory service.
General skillResearch and knowledgeModel Context Protocol
resourceful
@pedroreisper·Python
Exhaust a five-tier resource ladder before Claude says it cannot answer.
Project overview
This skill makes Claude inspect local files, system and Git history, authenticated MCPs, current web documentation, and delegated agents before giving up on a research-shaped question. A Stop hook blocks premature hedging unless enough tiers were tried or a structured escalation report is present. Its tier detection is heuristic and its phrase catalog mainly covers English and Portuguese.
Keeps Claude’s prose more natural across a session, with skeptic passes and dash cleanup.
Project overview
naturalize uses a SessionStart hook to nudge Claude toward natural prose throughout a session while skipping code, commits, and file paths. Each run rewrites the draft, rereads it as a skeptic, fixes passages that still sound synthetic, and ends by removing em dashes and en dashes. You can install it as an on-demand skill instead, and its test corpus covers the cataloged patterns.
Tracks attention strain and escalates from nudges to enforced cooldowns.
Project overview
Burnout Guard estimates burnout risk from self-report plus behavioral signals such as prompt heartbeats, late-night work, continuous blocks, and active-day streaks. Six levels range from quiet monitoring to single-task mode and a platform-level lockout with a cooldown, and Claude Code hooks can enforce the decision before a prompt reaches the model. State stays local, optional sleep and calendar inputs affect strain, and the project explicitly is not a medical device.
Checks proposed dependency versions against live OSV and CVE data before the agent writes them.
Project overview
safedeps grounds dependency choices in current OSV vulnerability records instead of a model’s frozen training knowledge. When an agent is about to add a package and version, it checks for known flaws and can stop the suggestion before it enters the project. This is a narrow decision-time guard, not a replacement for npm audit, pip-audit, osv-scanner, or a broader CI security scan.
Turn premature “can’t” replies into constraints, options, tradeoffs, and a recommendation.
Project overview
This Claude Code plugin intercepts assistant replies containing can’t or cannot and asks the model to name the actual constraint, list feasible alternatives, explain tradeoffs, and recommend a path forward. It is intentionally narrow—a Stop hook plus a bundled reframe skill—and may misfire on genuine safety or legal refusals.
Enforce scope, safety checks, and audit evidence for coding agents.
Project overview
Code-Warden adds verifiable controls around delegated coding work: declared scope, secret and file-size checks, runtime hooks, a Git pre-commit backstop, CI reports, and audit receipts. Teams can adopt the layers independently and use baselines for existing codebases. It is not a sandbox, and Codex exposes fewer hook surfaces than Claude Code.
Manage reminders in natural language and show due items at session start.
Project overview
Add, list, complete, and review reminders through a natural-language command, with records stored in a local JSON file. A SessionStart hook shows overdue, due-today, and upcoming items when Claude Code opens, while completed reminders remain available in history. It uses Bash and Python 3 without external dependencies.
Keep an Obsidian project vault aligned with code changes made by Claude.
Project overview
Claude Autopilot for Obsidian makes Claude read relevant notes before editing code, update the vault alongside the change, archive sessions, cross-link related work, and optionally sync through git. The pure-Python toolkit runs on macOS, Linux, and Windows; archives stay local and are secret-redacted by default, while pushing them requires explicit opt-in.
Connect Claude Code lifecycle events to macOS Lights traffic signals.
Project overview
lights-hooks merges Claude Code lifecycle hooks into `~/.claude/settings.json` so the macOS Lights app can show red while work runs, yellow for permission states, and green when idle. It saves a timestamped backup before changing existing hooks and removes only its own port-9876 entries on uninstall. The setup requires macOS, Lights.app running locally, and Claude Code.
Prompt coding agents to run the minimum relevant checks before every git push.
Project overview
Claude Think Twice combines a reflection skill with a Claude Code pre-push hook. When a push is attempted, it inspects the pending file types, presents the corresponding minimum scan matrix, and asks the human whether to proceed, helping expose rushed assumptions before CI. The hook cannot verify that scans were actually run and is not a linter, test runner, or replacement for CI; its enforcement is the human confirmation moment.
Keep macOS system proxy settings synchronized with shells and GUI apps.
Project overview
mac-proxy-sync reads the active macOS proxy from scutil and exports common HTTP, HTTPS, SOCKS, and no-proxy variables whenever zsh starts. A prompt hook refreshes the values at most once per minute so long-running sessions follow rotating Clash, V2Ray, or Mihomo ports, while syncproxy mirrors them into launchctl for GUI subprocesses. It also unsets stale values when the system proxy is disabled and is specific to macOS and zsh.