dotskills
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 11 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/vincentkoc/dotskills with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/vincentkoc/dotskills'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/vincentkoc--dotskills)HTML
<a href="/listings/vincentkoc--dotskills"><img src="/api/listings/vincentkoc--dotskills/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Install one skill:
npx skills add vincentkoc/dotskills --skill ghcrawl-cluster-operator -y npx skills add vincentkoc/dotskills --skill agent-device-lab -y npx skills add vincentkoc/dotskills --skill codebase-memory-mcp -y npx skills add vincentkoc/dotskills --skill codex-goal-mining -y npx skills add vincentkoc/dotskills --skill codex-session-recovery -y npx skills add vincentkoc/dotskills --skill crabpot-perf-metrics -y npx skills add vincentkoc/dotskills --skill technical-deslop -y npx skills add vincentkoc/dotskills --skill technical-documentation -y npx skills add vincentkoc/dotskills --skill session-done -y npx skills add vincentkoc/dotskills --skill technical-integrations -y npx skills add vincentkoc/dotskills --skill technical-skill-finder -y npx skills add vincentkoc/dotskills --skill openclaw-github-dedupe -y npx skills add vincentkoc/dotskills --skill openclaw-pr-batch-sweep -y npx skills add vincentkoc/dotskills --skill org-branch-cleanup -y npx skills add vincentkoc/dotskills --skill openclaw-bun-upgrade -y npx skills add vincentkoc/dotskills --skill operations-worktree -y npx skills add vincentkoc/dotskills --skill crawlkit -y npx skills add vincentkoc/dotskills --skill graincrawl -y npx skills add vincentkoc/dotskills --skill github-airplane-mode -y npx skills add vincentkoc/dotskills --skill semantic-slicing -y npx skills add vincentkoc/dotskills --skill tmux-agent-lane-orchestrator -y npx skills add vincentkoc/dotskills --skill tmux-lane-orchestrator -y
List available public skills:
npx skills add vincentkoc/dotskills --list
`github-airplane-mode` requires an already-installed compatible `low-data` command. Installing the skill does not install a network guard or change its mode.
Default sync targets are managed automatically by vercel skills.
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 144 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- {
- "name": "dotskills",
- "owner": {
- "name": "vincentkoc"
- },
- "metadata": {
- "description": "A .skills collection: prompts with resources and scripts as reusable AI runtime modules.",
- "version": "1.0.0"
- },
- "plugins": [
- {
- "name": "agent-device-lab",
- "source": "./skills/agent-device-lab",
- "skills": "./",
- "description": "Prepare a Mac and one physical Android or iPhone for an authorized test."
- },
- {
- "name": "codebase-memory-mcp",
- "source": "./skills/codebase-memory-mcp",
- "skills": "./",
- "description": "Bring up `codebase-memory-mcp` for the owning Git checkout and prove the graph is usable before relying on it for code discovery. Keep linked worktrees on the owner's graph instead of creating one graph per branch."
- },
- {
- "name": "codex-goal-mining",
- "source": "./skills/codex-goal-mining",
- "skills": "./",
- "description": "Recover structured Codex goal history and convert free-form runs into evidence-backed, repeatable suites."
- },
- {
- "name": "codex-session-recovery",
- "source": "./skills/codex-session-recovery",
- "skills": "./",
- "description": "Recover crashed, detached, or confusing agent sessions while preserving the best restore source."
- },
- {
- "name": "crabpot-perf-metrics",
- "source": "./skills/crabpot-perf-metrics",
- "skills": "./",
- "description": "Turn Crabpot/OpenClaw perf artifacts into a clear read: what changed, what is real signal, what is stale/noisy, and what action is worth taking."
- },
- {
- "name": "crawlkit",
- "source": "./skills/crawlkit",
- "skills": "./",
- "description": "Maintain crawlkit, the shared Go library for crawl app archive infrastructure."
- },
- {
- "name": "ghcrawl-cluster-operator",
- "source": "./skills/ghcrawl-cluster-operator",
- "skills": "./",
- "description": "Operate ghcrawl as a local-first GitHub issue and pull request crawler: inspect the SQLite store, pull GitHub data, refresh summaries and embeddings, build clusters, and extract cluster evidence through deterministic CLI commands."
- },
- {
- "name": "github-airplane-mode",
- "source": "./skills/github-airplane-mode",
- "skills": "./",
- "description": "Switch an already-installed local Git and GitHub download guard."
- },
- {
- "name": "graincrawl",
- "source": "./skills/graincrawl",
- "skills": "./",
- "description": "Maintain `graincrawl`, the Go CLI that archives Granola notes, transcripts,"
- },
- {
- "name": "openclaw-bun-upgrade",
- "source": "./skills/openclaw-bun-upgrade",
- "skills": "./",
- "description": "Move an authorized Linux OpenClaw service to one verified Bun artifact while preserving"
- },
- {
- "name": "openclaw-github-dedupe",
- "source": "./skills/openclaw-github-dedupe",
- "skills": "./",
- "description": "Use this skill when a cluster of GitHub issues and pull requests has been reported for a common failure mode (Slack, iMessage, support threads, or manual list), and you need an evidence-based dedupe recommendation, execution pass, or autonomous closeout run."
- },
- {
- "name": "openclaw-pr-batch-sweep",
- "source": "./skills/openclaw-pr-batch-sweep",
- "skills": "./",
- "description": "Drive a continuing queue of real, low-risk OpenClaw contributor bug fixes through qualification, repair, proof, and landing. Work in batches of up to 20 without padding the batch with micro-patches, speculative cleanup, or risky surfaces."
- },
- {
- "name": "operations-worktree",
- "source": "./skills/operations-worktree",
- "skills": "./",
- "description": "Create and manage worktrees safely and consistently across projects while avoiding stale branch bases."
- },
- {
- "name": "org-branch-cleanup",
- "source": "./skills/org-branch-cleanup",
- "skills": "./",
- "description": "Remove organization branch clutter without guessing which refs are safe to delete."
- },
- {
- "name": "semantic-slicing",
- "source": "./skills/semantic-slicing",
- "skills": "./",
- "description": "Turn a large repo into reviewable semantic slices with evidence. Use code shape, threat candidates, issue clusters, and support chatter together so review budget lands on the right parts of the system."
- },
- {
- "name": "session-done",
- "source": "./skills/session-done",
- "skills": "./",
- "description": "Capture the outcome, evidence, unresolved work, and next action without making session completion a cleanup or memory-write command."
- },
- {
- "name": "technical-deslop",
- "source": "./skills/technical-deslop",
- "skills": "./",
- "description": "Clean branch/staged diffs by removing AI-style slop and restoring repo-consistent style without changing behavior."
- },
- {
- "name": "technical-documentation",
- "source": "./skills/technical-documentation",
- "skills": "./",
- "description": "Produce, review, and audit technical documentation that is clear, unambiguous, and maintainable for both humans and agents. Covers product docs, contributor-governance files, and agent instruction files. Prose follows Simplified Technical English so a reader or an agent cannot misread it. Audits scale to trees with hundreds of pages through sharding, adversarial checks, and parallel remediation."
- },
- {
- "name": "technical-integrations",
- "source": "./skills/technical-integrations",
- "skills": "./",
- "description": "Design and review external-vendor integrations using existing internal patterns, with vendor/framework-agnostic workflows for API, RFC, SDK, and rollout planning."
- },
- {
- "name": "technical-skill-finder",
- "source": "./skills/technical-skill-finder",
- "skills": "./",
- "description": "Find recurring pain points from local agent logs and convert them into actionable skill candidates, reuse opportunities, or existing skill updates."
- },
- {
- "name": "tmux-agent-lane-orchestrator",
- "source": "./skills/tmux-agent-lane-orchestrator",
- "skills": "./",
- "description": "Turn a tmux window of coding-agent workers into a visible, auditable lane with clear state and next actions."
- },
- {
- "name": "tmux-lane-orchestrator",
- "source": "./skills/tmux-lane-orchestrator",
- "skills": "./",
- "description": "Act as the manager for one tmux worker lane. Keep visibility over the lane's panes, infer what each worker is doing from tmux and Codex logs, surface blockers, and give the operator short status summaries with concrete next actions."
- }
- ]
- }
README.md · README · 211 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- <div align="center">
- 
- # .skills (dotskills)
- [](https://github.com/vincentkoc/dotskills/actions/workflows/validate.yml)
- [](https://github.com/vincentkoc/dotskills/actions/workflows/release.yml)
- [](https://github.com/vincentkoc/dotskills/actions/workflows/verify-install.yml)
- [](https://github.com/vincentkoc/dotskills/actions/workflows/codeql.yml)
- [](LICENSE)
- [](https://github.com/vincentkoc/dotskills/tags)
- [](CONTRIBUTING.md)
- </div>
- We are moving from "prompt as text" to **skill as runtime module**. Each skill acts like a lightweight, containerized application for AI work: a stable interface, opinionated workflow, and bundled resources that can be installed, versioned, tested, and reused across projects.
- ## How this repo works
- This is my personal **.skills** repository for Codex, Cursor, OpenClaw and agent-first tooling. `.skills` is the dotfiles mindset applied to AI execution: instead of one-off prompts, this repo stores reusable skill units that bundle:
- - prompt logic (`SKILL.md` / `AGENT.md` / `AGENTS.md`)
- - references and knowledge assets
- - scripts for deterministic execution
- - repeatable validation + publishing workflows
- - support external skills through submodules
- - git managed personal registry and github action hooks
- ## Configuration ownership
- Keep each configuration with the component that installs and validates it:
- | Configuration | Owner |
- |---|---|
- | Reusable skills, references, and bundled helpers | This repository |
- | Personal global agent instructions | [dotfiles `.codex/AGENTS.md`](https://github.com/vincentkoc/dotfiles/blob/master/.codex/AGENTS.md) |
- | Repository-specific `AGENTS.md` instructions | The repository they govern |
- | MCP registrations, machine-local settings, and credential injection | The platform's dotfiles/bootstrap installer |
- `make sync` selects agent targets such as `codex` and `cursor`. It installs
- skills and the supported slash-command prompts; it does not render repository
- instructions or MCP configuration. Use the platform installer to reconcile a
- machine's MCP setup, and keep credentials out of skill packages. This avoids a
- second configuration generator competing with those owners.
- ## Public skills
- | Skill | What it does | Install |
- |---|---|---|
- | `codebase-memory-mcp` | Index canonical Git checkouts through a guarded CLI and audit duplicate worktree caches. | `npx skills add vincentkoc/dotskills --skill codebase-memory-mcp -y` |
- | `agent-device-lab` | Prepare physical Android and iPhone tests, with separate iPhone Mirroring setup. | `npx skills add vincentkoc/dotskills --skill agent-device-lab -y` |
- | `codex-goal-mining` | Mine local or fleet Codex goal history into evidence-backed rerun suites. | `npx skills add vincentkoc/dotskills --skill codex-goal-mining -y` |
- | `codex-session-recovery` | Recover Codex and Claude tmux sessions without overwriting restore evidence. | `npx skills add vincentkoc/dotskills --skill codex-session-recovery -y` |
- | `crabpot-perf-metrics` | Interpret Crabpot and OpenClaw performance artifacts without over-reading noise. | `npx skills add vincentkoc/dotskills --skill crabpot-perf-metrics -y` |
- | `crawlkit` | Maintain and release the shared crawl app Go library. | `npx skills add vincentkoc/dotskills --skill crawlkit -y` |
- | `graincrawl` | Maintain and release the local-first Granola archive CLI. | `npx skills add vincentkoc/dotskills --skill graincrawl -y` |
- | `github-airplane-mode` | Switch an already-installed compatible local Git/GitHub download guard. | `npx skills add vincentkoc/dotskills --skill github-airplane-mode -y` |
- | `ghcrawl-cluster-operator` | Operate ghcrawl local sync, clustering, and cluster inspection from the CLI. | `npx skills add vincentkoc/dotskills --skill ghcrawl-cluster-operator -y` |
- | `openclaw-bun-upgrade` | Qualify and promote pinned Bun fork builds through the existing service owner. | `npx skills add vincentkoc/dotskills --skill openclaw-bun-upgrade -y` |
- | `openclaw-github-dedupe` | Triage GitHub issue/PR clusters across repositories, preserving contributor credit while selecting canonical fixes and safe closures. | `npx skills add vincentkoc/dotskills --skill openclaw-github-dedupe -y` |
- | `openclaw-pr-batch-sweep` | Select and land bounded batches of low-risk OpenClaw contributor pull requests. | `npx skills add vincentkoc/dotskills --skill openclaw-pr-batch-sweep -y` |
- | `org-branch-cleanup` | Audit and safely prune stale merged branches across a GitHub organization. | `npx skills add vincentkoc/dotskills --skill org-branch-cleanup -y` |
- | `operations-worktree` | Create safe git worktrees from fresh remote defaults instead of stale local branches. | `npx skills add vincentkoc/dotskills --skill operations-worktree -y` |
- | `semantic-slicing` | Build local feature, threat, issue, and support maps for focused repo review. | `npx skills add vincentkoc/dotskills --skill semantic-slicing -y` |
- | `technical-deslop` | Ship clean diffs fast: remove AI noise and keep behavior unchanged. | `npx skills add vincentkoc/dotskills --skill technical-deslop -y` |
- | `technical-documentation` | Build, review, and audit docs with Simplified Technical English, sharded large-tree audits, and Workflow/sub-agent fan-out. | `npx skills add vincentkoc/dotskills --skill technical-documentation -y` |
- | `session-done` | Prepare a session handoff; retain a Markdown note only when requested. | `npx skills add vincentkoc/dotskills --skill session-done -y` |
- | `technical-integrations` | Design integrations that land: vendor‑agnostic API/RFC/SDK plans with rollout safety. | `npx skills add vincentkoc/dotskills --skill technical-integrations -y` |
- | `technical-skill-finder` | Turn real agent pain into new skills: mine logs, rank wins, draft next steps. | `npx skills add vincentkoc/dotskills --skill technical-skill-finder -y` |
- | `tmux-agent-lane-orchestrator` | Monitor tmux coding-agent lanes and produce evidence-backed manager summaries. | `npx skills add vincentkoc/dotskills --skill tmux-agent-lane-orchestrator -y` |
- | `tmux-lane-orchestrator` | Coordinate one tmux worker lane from its matching manager pane. | `npx skills add vincentkoc/dotskills --skill tmux-lane-orchestrator -y` |
- Local-only skills may live under ignored `private-skills/`, but must never be tracked in this public repository.
- ## Install
- Install one skill:
- ```bash
- npx skills add vincentkoc/dotskills --skill ghcrawl-cluster-operator -y
- npx skills add vincentkoc/dotskills --skill agent-device-lab -y
- npx skills add vincentkoc/dotskills --skill codebase-memory-mcp -y
- npx skills add vincentkoc/dotskills --skill codex-goal-mining -y
- npx skills add vincentkoc/dotskills --skill codex-session-recovery -y
- npx skills add vincentkoc/dotskills --skill crabpot-perf-metrics -y
- npx skills add vincentkoc/dotskills --skill technical-deslop -y
- npx skills add vincentkoc/dotskills --skill technical-documentation -y
- npx skills add vincentkoc/dotskills --skill session-done -y
- npx skills add vincentkoc/dotskills --skill technical-integrations -y
- npx skills add vincentkoc/dotskills --skill technical-skill-finder -y
- npx skills add vincentkoc/dotskills --skill openclaw-github-dedupe -y
- npx skills add vincentkoc/dotskills --skill openclaw-pr-batch-sweep -y
- npx skills add vincentkoc/dotskills --skill org-branch-cleanup -y
- npx skills add vincentkoc/dotskills --skill openclaw-bun-upgrade -y
- npx skills add vincentkoc/dotskills --skill operations-worktree -y
- npx skills add vincentkoc/dotskills --skill crawlkit -y
- npx skills add vincentkoc/dotskills --skill graincrawl -y
- npx skills add vincentkoc/dotskills --skill github-airplane-mode -y
- npx skills add vincentkoc/dotskills --skill semantic-slicing -y
- npx skills add vincentkoc/dotskills --skill tmux-agent-lane-orchestrator -y
- npx skills add vincentkoc/dotskills --skill tmux-lane-orchestrator -y
- ```
- List available public skills:
- ```bash
- npx skills add vincentkoc/dotskills --list
- ```
- `github-airplane-mode` requires an already-installed compatible `low-data`
- command. Installing the skill does not install a network guard or change its mode.
- Default sync targets are managed automatically by vercel skills.
- ## Codex slash commands
- This repo can also mirror Codex slash-command prompts from `vendor/` and sync them to `~/.codex/prompts`.
- - `make sync` now syncs skills and slash commands for the `codex` profile.
- - vendored command currently included: `/sectriage` from `steipete/agent-scripts`.
- Local `make sync` and `make sync-copy` require macOS or Linux. Sync updates only
- same-source symlinks or unchanged managed copies. An older unmarked copy can be
- adopted when all content and modes match the current source. Different copies,
- extra local files, and foreign links require reconciliation before sync. Failed
- publication never overwrites an intervening destination; the error names any
- retained recovery directory that needs review before retrying.
- ### Move one managed skill to a new source
- Ordinary sync never infers a source change from a skill name. After verifying
- both sources' ownership and revisions, transition one unchanged managed copy:
- ```bash
- python3 scripts/managed_sync.py install-directory \
- --source /absolute/new-source/skill \
- --previous-source /absolute/old-source/skill \
- --destination /absolute/installed/skill --mode copy
- ```
- The old source must still exist and match the installed marker's binding.
- The installed content and modes must still match its recorded digest. Unmarked
- copies, symlinks, missing destinations, local edits, extra files, and wrong old
- bindings are refused. This operation copies the complete new skill, including
- relative resources, and publishes its new source binding using the same staging,
- no-replace, and recovery checks as ordinary sync. It does not alter either source
- or sibling skills. No bulk-sync option, marker reset, or trust decision is implied.
- On success, the command reports the retained `.agent-skills-stage-*` directory
- on stderr. Its `previous/` directory holds the old installed copy and marker.
- An open file descriptor can still write to that old copy after final validation;
- another digest check cannot make deletion safe. Only explicit source transitions
- retain successful recovery this way; ordinary sync cleanup is unchanged. Inspect
- and reconcile the retained copy through the existing recovery owner, after old
- writers have finished. Success does not authorize deleting that recovery.
- Record the old and new source revisions in the existing task/change record before
- running it. Keep immutable release sources intact; do not edit or repin a whole
- release to update one skill. After success, use the new source for ordinary sync;
- replaying the old binding is refused. A failed publication may also retain recovery
- as described above: inspect that recovery before any retry, never remove markers
- to force adoption. Source trust or release-ownership changes still require their
- own authorization; this explicit operation is not a bypass.
- ## Repository layout
- ```text
- skills/ # First-party skills
- <skill-name>/
- SKILL.md|AGENT.md|AGENTS.md
- references/
- scripts/
- private-skills/ # Private/local-only skills (not published)
- <skill-name>/
- SKILL.md|AGENT.md|AGENTS.md
- references/
- scripts/
- vendor/ # Third-party mirrored/imported skills
- .assets/ # Shared visual assets (banner/icon)
- bin/agent-skills # List/validate/sync/import
- scripts/ # Validation + publishing automation
- catalog.yaml # Skill metadata catalog
- .claude-plugin/marketplace.json
- releases/skills.json
- ```
- ## Validation
- `make validate` now runs:
- 1. repo policy checks (`Purpose/When to use/Workflow/Inputs/Outputs`, entry-file checks)
- 2. AgentSkills spec checks for local `SKILL.md` skills (`name`/`description` constraints, frontmatter validity, relative resource references, one-level reference depth)
- 3. public skill OpenAI metadata checks (`agents/openai.yaml`, required interface fields, default icons/brand from `AGENTS.md`, and icon path existence)
- If installed, the external reference validator is also used:
- ```bash
- skills-ref validate ./skills/<skill-name>
- ```
- ## Why .skills
- Dotfiles configure machines.
- `.skills` configures AI execution quality.
- This repo is meant to be composable, auditable, and practical: skills should be testable artifacts, not throwaway prompt snippets.
- ## Contributing
- Read [CONTRIBUTING.md](CONTRIBUTING.md) and [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md).
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.