Depth Lock
Lock scope, review rounds and wall-clock checkpoints before non-trivial work, and send everything outside the task to follow-ups. Use at the start of any plan, feature, fix train or multi-agent workflow, when a review loop starts repeating, when findings drift beyond the task, or when the operator says "не полируй", "не растягивай", "полировка круглого камня", "stop polishing" or "timebox".
Install
npx skills add malakhov-dmitrii/depth-lock --skill depth-lockgit clone --depth 1 https://github.com/malakhov-dmitrii/depth-lock.git /tmp/depth-lock
mkdir -p ~/.claude/skills
cp -r /tmp/depth-lock/. ~/.claude/skills/depth-lockPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Depth lock Finish the task at the agreed depth, fast. Robust means the safety invariants hold. It does not mean endless review rounds or cleanup of neighbouring code. ## 1. Before building: write the scope lock into the task journal Use the active task journal (for example `.agent/tasks/<task>.md`). If there is no journal, put it in the plan file. ``` ## Scope lock Goal: <one user-visible sentence> Acceptance: <testable bullets: what proves it works> In scope: <items or conditions> Non-goals: <explicit list, including tempting refactors> Blockers (the only findings that stop the ship): <project list; default: duplicate action (double send, charge or write), lost data, stuck state with no recovery, security hole, broken production path> Review caps: plan ≤ 2 rounds, code ≤ 2 rounds; every round after the first reports blockers only Checkpoints (wall clock, UTC): plan HH:MM · build HH:MM · gate HH:MM · ship HH:MM Follow-ups: <journal section, issue tracker or spawned tasks> ``` - Estimate in agent wall-clock time with parallel agents, not in human-team days. - If the scope is clear, state the lock and proceed. Ask the operator only when the scope is genuinely ambiguous. ## 2. During the work - Classify every review finding as `blocking`, `deferred` or `approved-expansion`. Only `blocking` findings and approved expansions enter the current work. - A later review round gets the prompt "blockers only; no style, naming, nice-to-haves or scope expansion".
Details
- Skill name
depth-lock- Repository
- malakhov-dmitrii/depth-lock
- License
- MIT
- Repo updated
- 1 October 2026
FAQ
What does the Depth Lock skill do?
Lock scope, review rounds and wall-clock checkpoints before non-trivial work, and send everything outside the task to follow-ups. Use at the start of any plan, feature, fix train or multi-agent workflow, when a review loop starts repeating, when findings drift beyond the task, or when the operator says "не полируй", "не растягивай", "полировка круглого камня", "stop polishing" or "timebox".
How do I install the Depth Lock skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add malakhov-dmitrii/depth-lock --skill depth-lock. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Depth Lock popular?
It lives in malakhov-dmitrii/depth-lock, which has 5 GitHub stars and 1 forks.
Related skills
Job Application Tracker
5daliaabbruciati/job-application-tracker-SKILL
Automatically syncs Gmail and Google Calendar with a Notion database to keep track of the status of job applications the user has submitted. Use this skill whenever the user asks to "sync my job applications," "update my job tracker," "check where my interviews stand," or something similar — even if they don't explicitly mention Notion, Gmail, or Calendar, and even if they don't have a Notion database set up for this yet. Requires (or guides the user through setting up) the Notion, Gmail, and Google Calendar MCP connectors.
Workflow Herdr
5letya999/workflow-herdr
Start a Herdr work run for a change. Use when starting work, doing a task in Herdr, splitting work across panes, or chatting from a Herdr terminal - even if they just say do this. Do NOT use for a solo one-pane edit, headless delegates, or product job queues.
Memory Management
4jau123/claude-memory-manager
Claude Code auto-memory 与 CLAUDE.md 的管理方法 — 判断值不值得记、记到哪一层(行内注释 / .claude/rules / memory / CLAUDE.md)、怎么写 frontmatter 与 description、何时 update 何时拆分、索引 MEMORY.md 的容量与 hub 化。用户说"记一下""新增/更新记忆""沉淀经验""复盘""改 memory schema""audit memory""记忆系统乱了"时使用;功能开发或长 debug 收尾有值得沉淀的结论时也用。
Lecture Notes
6orhoncan/lecture-notes-skill
Generate comprehensive undergraduate lecture notes for social sciences (especially economics) with mathematical foundations, daily-life examples, and practice questions. Use when user asks to create, generate, or write lecture notes, lecture materials, or teaching content. Supports Turkish and English output, Obsidian-compatible markdown with LaTeX/TikZ, and optional syllabus integration.
Obsidian Skill
5gmickel/obsidian-skill
Use Obsidian's official CLI to work with a running Obsidian app and its configured vaults. Use for live workspace state, indexed search and metadata, daily notes, tasks, links and backlinks, Bases, templates, bookmarks, plugins, themes, Sync status, and developer inspection. Prefer ordinary filesystem tools when the task only needs direct file reads, writes, listing, or text search.
Email HTML QA
6EmailBoutique-Digital-Inc/email-html-qa-skill
Opinionated, pre-send QA for HTML email. Use when reviewing, auditing, or QA-ing HTML email code before it ships — checking rendering across clients (Outlook Windows/Mac, Gmail, Apple Mail/iOS), forced dark mode, responsive / mobile, code size and Gmail clipping, accessibility mechanics, images, links/CTAs, preheader, personalization fallbacks, and that client-specified compliance elements are present and functional. Also reads and reports the authentication results (SPF/DKIM/DMARC) and unsubscribe headers from a received message. Reports findings by severity and diagnoses before changing code
Browse the most popular Claude skills, all productivity & notes skills, or write your own with the skills generator.