ops-mode
Activate full operational constraints mode for disciplined, efficient execution. Use when the user asks for ops mode.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-11
- Stars
- ★ 6
- Forks
- 1
01 前言 Front Matter
- name
- ops-mode
- description
- Activate full operational constraints mode for disciplined, efficient execution. Use when the user asks for ops mode.
03 SKILL.md 正文
Operational Mode Activated
This session is now operating in OPS-MODE with comprehensive operational constraints enforced.
Active Constraints
Token & Context Efficiency
- Minimize context loading - only essential information
- Compress outputs - no verbose explanations unless requested
- Focused reasoning - minimal "thinking out loud"
- When approaching 70% context, initiate compaction
Tool Protocols
Documentation & Research:
- Context7 MCP for library/API documentation (retry once on failure, then fallback)
- WebSearch for current information beyond training cutoff
- WebFetch for specific documentation pages
Code Search:
| Need | Tool |
|---|---|
| Text patterns | GrepTool |
| Code structures | Skill(ast-grep) |
| Complex analysis | Python script |
| File patterns | Glob |
| Exploration | Task(Explore) |
Parallel Execution Mandate
Before any multi-step task, evaluate:
Independent subtasks exist?
├── NO → Sequential execution
└── YES → Dependencies between them?
├── YES → Parallelize independent portions
└── NO → DISPATCH PARALLEL AGENTS NOW
Use multiple Task tool invocations in a single message for parallelization.
Development Discipline
Principles enforced:
- KISS - Keep It Simple, Stupid
- DRY - Don't Repeat Yourself
- SOLID - Object-oriented design principles
- SOC - Separation of Concerns
- 12 Factor App - Application architecture
TDD Mandate:
- RED → GREEN → REFACTOR cycle
- Write test BEFORE implementation
- No rationalizations for skipping
Impact Analysis:
- Use Skill(impact-analysis) for ALL code changes
- HIGH risk impacts must be resolved before proceeding
System Mapping Pre-Flight
Before modifying code, perform 5-point check (if applicable):
- Architecture - Files, modules, entry points
- Execution Flow - Call chains, async boundaries
- Logging - Existing observability
- Dependencies - Imports, external services
- Environment - Config, env vars, flags
Use Skill(system-mapping) for detailed protocol.
Planning Output Protocol
At end of any planning:
- Explicitly ask for approval
- Offer options: proceed / create AA-MA docs / adjust
- Identify single next action
Session Start: Lessons Review
Before executing any task, check if the project has a docs/lessons.md file. If it exists, read the top 10 entries and internalize them. If it doesn't exist, note this and continue.
This enforces the self-improvement-loop rule: "At the beginning of EVERY new session, your first operational action must be to review lessons."
Pre-Task Checklist
Before proceeding with your task:
- [ ] Lessons reviewed - Read
docs/lessons.mdtop entries (if exists) - [ ] Skills reviewed - Checked available skills for relevance
- [ ] Parallelization evaluated - Identified if subagents can help
- [ ] System mapping needed? - For code changes to unfamiliar areas
- [ ] TDD applicable? - Plan tests first if implementing code
- [ ] AA-MA active? - Check
.claude/dev/active/for existing plans
Quick Commands
| Command | Purpose |
|---|---|
Skill(operational-constraints) |
Full constraint documentation |
Skill(system-mapping) |
5-point pre-flight protocol |
Skill(test-driven-development) |
TDD workflow |
Skill(impact-analysis) |
Change risk assessment |
Skill(dispatching-parallel-agents) |
Parallel execution patterns |
Confirmation
OPS-MODE: ACTIVE
Constraints enforced:
✓ Lessons review (self-improvement loop)
✓ Token efficiency
✓ Tool protocols (Context7, GrepTool, ast-grep)
✓ Parallel execution evaluation
✓ TDD discipline
✓ System mapping pre-flight
✓ Planning approval protocol
Session ready for disciplined execution.
What task would you like to work on?
04 同仓库更多 Skills
deploy-helper
You can use this to help with deploy stuff, deploying, deployments, releases, shipping, pushing to prod, going live, rollouts.
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
verify-plan
Slash entry for plan verification: runs plan-verification on an existing AA-MA plan, writes [task]-verification.md, and with --iterate loops until the plan converges. Use when the user types /verify-plan or wants the verification report saved.
verify-impl
Post-impl adversarial review symmetric to /verify-plan. Dispatches up to 5 parallel audit agents (code-reviewer, security-auditor, tdd-sequence-auditor, context7-evidence-auditor, future-proofing-auditor) based on the milestone's plan-declared Audit-Profile. CRITICAL findings surface via AskUserQuestion accept/dispute/defer panel before §7.3 user authorization. Introduced in v0.8.0 per ADR-0005. Invoked by Phase 6.8 of /execute-aa-ma-milestone.
understand-codebase
Onboard to a new, inherited, or shared codebase. Read it, understand it, map it, and learn its conventions, versioning, tests, tech stack, rules (AGENTS.md / CLAUDE.md / .cursorrules / etc.), build/run/CI, integrations, security posture, and repo health — then produce an honest pros/cons verdict, a "contribute safely" playbook, and an "add a feature" playbook, written to ONBOARDING.md at the repo root plus a .claude/onboarding/ set of deep-dives — and optionally author an AGENTS.md if one is missing, or review and propose improvements to an existing one. Tiered (Quick / Standard / Deep). Reuses codemem (index + diagrams), gsd-map-codebase, system-mapping, code-intelligence, impact-analysis rather than re-implementing them; Deep tier runs a TeamCreate agent-team. Keywords: new codebase, shared codebase, inherited code, onboard, understand this repo, how do I contribute, how do I add a feature, ramp up, get oriented, ONBOARDING.md, AGENTS.md, codebase walkthrough, joining a project.
token-compression
Output token compression for AA-MA workflows. Reduces response tokens by 40-75% while preserving full technical accuracy. Three intensity levels mapped to execution modes: lite (HITL), full (general), ultra (AFK). Auto-activates during AA-MA execution based on task Mode field. Manual activation via /compress lite|full|ultra.