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.
npx agentmods add instructions/igoraguiar/bun-pg-mcp/agents-mdgit clone --depth 1 https://github.com/igoraguiar/bun-pg-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/igoraguiar/bun-pg-mcp/agents-md)<a href="https://agentmods.dev/instructions/igoraguiar/bun-pg-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/igoraguiar/bun-pg-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00725 | $0.00725 |
| Opus 5 | $0.00362 | $0.00362 |
| Sonnet 5 | $0.00145 | $0.00145 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
bun-pg-mcp AGENTS.md 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project Context
pg-mcp is a PostgreSQL Model Context Protocol server that enables AI coding assistants to query, inspect, and manage multiple PostgreSQL databases securely via stdio, HTTP, or CLI modes. It is built for developers using tools like Cursor and Claude who need programmatic database access with auto-reloading multi-database configuration.
Tech Stack
- Runtime: Bun latest
- Language: TypeScript 5+ with strict mode (
tsconfig.json) - MCP SDK:
@modelcontextprotocol/sdk ^1.12.1 - Web: Express ^5.1.0 for HTTP mode
- Validation: Zod ^3.25.76
- DB: Bun SQL client with connection pooling
Directory Map
src/
index.ts # Entry point, mode routing stdio/http/cli
server.ts # MCP server factory, prompts & tools registration
config.ts # ConfigManager with auto-reload watcher
sql-pool.ts # Lazy connection pool with TTL reaper
handlers.ts # Tool handlers (pg_*, pg_db_*)
cli.ts # CLI argument parsing & execution
db/
types.ts # PgTableDetails etc
helpers.ts # SQL helpers for schema/table introspection
tests/ # bun:test integration tests
docs/agents/ # Issue tracker, triage labels, domain docs
dist/ out/ # Build outputs
Commands
bun install # Install dependencies
bun run src/index.ts # Run dev server (stdio mode)
bun run build # Build to dist/ with bun build
bun run compile # Compile standalone binary to out/pg-mcp
bun test # Run tests (bun:test)
# Lint/Format: no formatter configured; rely on tsc strict checks
Rules of Engagement
✅ Always:
- Use TypeScript strict mode, no
anywithout justification - Add JSDoc to exported functions/classes
- Validate all external input with Zod schemas
- Redact credentials in URLs before returning to clients
- Use async/await, never block the event loop
- Keep handlers pure; side effects via ConfigManager/SqlPool only
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.
- 5d ago First seen · 93 lines · 725 tokens per session scan A c4a5d28db92f
bun-pg-mcp AGENTS.md is an instructions file published in the GitHub repository igoraguiar/bun-pg-mcp (1 stars, last pushed 25d ago), licensed MIT. It adds 725 tokens to every session, about $0.0036 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.
Other instructions, from other repositories
supabase copilot-instructions.md
Copilot instructions for supabase/supabase, covering copilot code review instructions, review policy — read this first, confidence threshold, what not to comment on and what to comment on (priority order).
PgQue CLAUDE.md
Instructions for NikolayS/PgQue, covering claude.md -- pgque, project, naming convention, architecture and stack.
supabase CLAUDE.md
Claude Code instructions for supabase/supabase, a project described as: The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications.
counterpoise-ledger CLAUDE.md
Instructions for jeffjjohnston/counterpoise-ledger, covering claude.md, project overview, development commands, essential commands and testing.
rosetta local-db-grounding.instructions.md
In a checkout, the resolved ros-help.db is untrusted until verified; the latest CI release DB is the grounding source of truth.
rosetta database.instructions.md
Use when working on SQLite schema, database initialization, WAL mode, FTS5 triggers, or table definitions.