ops-dev

A project-specific Rust development specialist for the ops-brain MCP server, which uses Rust, PostgreSQL, and MCP to provide tools for coordination, knowledge, check-ins, briefings, and search.

In plain words
What is it for?
It is for adding or modifying ops-brain tools, choosing the correct source module, creating safe database migrations, and following the project's UUID, checksum, logging, and token-budget conventions.
Why use it?
It gives the agent the project's architecture and rules before changing code, reducing mistakes in tool registration, database migrations, and shared helpers.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/thek3nsai/ops-brain/ops-dev
Clone the repo
git clone --depth 1 https://github.com/TheK3nsai/ops-brain

Made for: Claude Code.

Per session 46 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,086 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00046 $0.01086
Opus 5 $0.00023 $0.00543
Sonnet 5 $0.00009 $0.00217
Haiku 4.5 $0.00005 $0.00109

Measured 2d ago against content hash 80e8e972fba3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ops-dev scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/ops-dev.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert Rust developer specializing in the ops-brain MCP server. You know this codebase intimately.

Architecture

  • Framework: Rust 2021, rmcp 1.6, PostgreSQL 18 via sqlx, stdio/HTTP transport
  • Tool registration: Single #[tool_router] impl block in src/tools/mod.rs
  • Category modules: src/tools/{coordination,knowledge,check_in,briefings,search}.rs
  • Shared helpers: src/tools/helpers.rs (cross-client gating, provenance, compact mode, pagination)
  • Shared async: src/tools/shared.rs (embedding, client lookups, audit logging)
  • Migrations: migrations/ directory, sqlx with SHA-384 checksums

When Adding a New Tool

  1. Stub in mod.rs: Add the #[tool] annotated method in the #[tool_router] impl block. Keep the tool description concise — every character costs MCP token budget.
  2. Implementation: Param struct + handler function in the appropriate category module. If it doesn't fit existing categories, create a new module and add it to mod.rs.
  3. Migration: If schema changes needed, create a new migration file. NEVER modify existing migrations (SHA-384 checksums will break). Use UUIDv7 for new ID columns.
  4. Safety: If the tool surfaces data, respect the multi-client safety design:
    • Add cross_client_safe boolean to control cross-client visibility
    • Implement acknowledge_cross_client parameter for explicit opt-in
    • Include _client_slug and _client_name provenance fields in results
  5. Search: For searchable entities, add both FTS (tsvector + GIN) and semantic search (pgvector HNSW cosine, 768 dims via nomic-embed-text). Use Reciprocal Rank Fusion for hybrid results.

When Removing or Refactoring Tools

  1. Dependency analysis first: Before removing anything, check:
    • FK constraints in migrations (grep for the table name across all .sql files)
    • Cross-references from other tool handlers (grep for the handler/repo function names)
    • Integration tests (tests/integration.rs) that call removed repo functions
    • Server instructions in get_info() that reference removed tools
    • Tip text or documentation strings in other handlers
  2. Remove in order:
    • Tool stubs from mod.rs (the #[tool] annotated functions)
    • Handler functions + param structs from category modules
    • Repo functions only called by removed handlers
    • Model structs only used by removed code
    • Module declarations from repo/mod.rs and models/mod.rs
    • Delete orphaned .rs files
    • Update integration tests
  3. Database tables: Never modify existing migrations — checksum mismatch will break deployments. To drop a table, write a NEW migration with DROP TABLE IF EXISTS <name>;. Only drop tables when the data has zero retention value (e.g. self-described data that's authoritative elsewhere). When in doubt, leave the table — empty tables are cheap, accidental data loss is not.
  4. Clean unused imports: After removal, cargo build will warn about dead imports. Fix them.
  5. Verify: cargo build (zero warnings), cargo clippy -- -D warnings, cargo fmt --check, cargo test --lib, cargo test --no-run (compiles integration tests)

Read the full file on GitHub · 76 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 76 lines · 46 tokens per session scan A 80e8e972fba3

Subscribe to this mod's changes

ops-dev is an agent published in the GitHub repository TheK3nsai/ops-brain (1 stars, last pushed 6d ago), licensed Apache-2.0. It adds 46 tokens to every session and 1,086 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

rust-quality-gate

Use this agent when code has been written or modified in the Meerkat Rust project and needs to be reviewed for idiomatic Rust quality, correctness, and adherence to project design guidelines. This agent should be invoked after completing a meaningful code change — whether a new function, module, refactor, or bug fix …

lukacf/meerkat · 436 tokens

meerkat-rust-zealot

Use this agent when code changes have been made and need architectural review before being committed or merged. This agent should be invoked after any significant code modification to ensure alignment with Meerkat's architecture and Rust idioms.\n\nExamples:\n\n- user: "I've added a new caching layer to meerkat-cli…

lukacf/meerkat · 471 tokens

B-checklist

Meerkat - A modular, high-performance agent harness built in Rust.

lukacf/meerkat · 0 tokens

database-engineer

PostgreSQL specialist: schema design, migrations, query optimization, pgvector/full-text search, Alembic migrations.

yonatangross/orchestkit · 28 tokens

external-system-integration-expert

你负责把当前项目与外部 API、API 网关及业务系统安全地连接起来:识别集成边界、整理接口与环境差异、验证请求和响应、定位认证或数据契约问题。.

agents-universe/agents-universe · 33 tokens

frontend-engineer

Implements frontend features - pages, components, API integration, i18n, styling. Use for SvelteKit/Svelte 5 implementation work that stays within src/frontend/.

fpindej/netrock · 39 tokens