Depth Lock — Development skill for Claude Code
Agent skill that locks scope, review rounds and wall-clock checkpoints, so coding agents ship instead of polishing forever.
How to install Depth Lock
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open malakhov-dmitrii/depth-lock and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Depth Lock does
Agent skill that locks scope, review rounds and wall-clock checkpoints, so coding agents ship instead of polishing forever.
Alternatives in Development
- Autoresearch — Claude Autoresearch Skill — Autonomous goal-directed iteration for Claude Code 2.5k ★
- /autopilot — Autonomous hunt loop with deterministic scope safety and configurable checkpoints 1.2k ★
- Add Transitions — Add professional scene transitions to your Remotion video 113 ★
README
depth-lock
An agent skill that stops coding agents from polishing a task forever.
Left alone, an agent will run a fourth review round "to be safe", tidy the files next to the change, turn a reviewer's nice-to-have into new work, and estimate "9–11 days" for something parallel agents finish in an afternoon. `depth-lock` makes it agree on the depth first and then hold it.
What it does
Before building, the agent writes a scope lock into the task journal or plan:
## Scope lock
Goal:
Acceptance:
In scope:
Non-goals:
Blockers: duplicate action, lost data, stuck state with no recovery, security hole, broken production path
Review caps: plan ≤ 2 rounds, code ≤ 2 rounds; rounds after the first report blockers only
Checkpoints (UTC): plan HH:MM · build HH:MM · gate HH:MM · ship HH:MM
Follow-ups:
While working, the agent:
- sorts every review finding into
blocking,deferredorapproved-expansion, and only blockers and approved expansions get worked on; - stops at two review rounds; blockers found in the last round get fixed test-first and checked in the gate, with no third round;
- cuts non-safety scope when a checkpoint slips and says what it cut;
- writes everything out of scope down as a follow-up (what, where, why, next step) instead of fixing it now or dropping it.
The close-out lists what was delivered against the acceptance list, what was cut, and where the follow-ups went.
The full rules are in [SKILL.md](SKILL.md).
Install
Claude Code:
git clone https://github.com/malakhov-dmitrii/depth-lock ~/.claude/skills/depth-lock
Codex and other agents that read `~/.agents/skills`:
git clone https://github.com/malakhov-dmitrii/depth-lock ~/.agents/skills/depth-lock
The skill triggers on plans, features, fix trains and multi-agent workflows, on review loops that keep repe
Related Skills
Pursue
Hand Kimi an objective to pursue AUTONOMOUSLY across multiple turns (experimental goal mode), bounded by a har
Talkthrough MCP
Don't write a bug report — record it. Local MCP: screen recording → transcript, frames, OCR, wall-clock eviden
Lane Capture
Capture and vision lane. Own passive screen capture and the frame-to-log wall-clock join that binds rendered t
Getting Better
A Claude Skill that front-loads clarifying questions, kills padded output, and locks in corrections as standin
Openscad Skill
Claude skill for OpenSCAD: parametric parts and multi-part assemblies, verified with numbers instead of pictur
Fde Simulation
Hands-on Forward Deployed Engineer role simulations: scope an enterprise engagement, build multi-agent prototy
Related Agents
Determinism Auditor
Hunts nondeterminism in the toolchain and in generated code — hash-ordered iteration, floats, wall clock, env,
Development Orchestrator Agent
Owns everything between contract lock and Gate 2 — branch verification, conditional shared-package-agent, para
Concurrency Reviewer
Reviews code changes for concurrency defects: check-then-act races, lost updates, transaction isolation assump