A procedure for testing the @inflexa-ai/harness library as an installed package rather than only testing its source files. It builds the package, links it into a small test project, and exercises it against a real PostgreSQL database running in Podman.

Use it after changing the harness to build its distribution files, import the public package, run its agent and tool interfaces, and test database-backed behavior.

It catches problems that appear only at the package boundary, such as missing exports or differences between source and built files. It checks the same kind of package use expected by the embedding command-line tool.

verify — Skill for Claude Code by inflexa-ai/inflexa

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.

Any agent

npx skills add inflexa-ai/inflexa --skill verify

Clone the repo

git clone --depth 1 https://github.com/inflexa-ai/inflexa

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/inflexa-ai/inflexa/verify/github.svg)](https://agentmods.dev/skills/inflexa-ai/inflexa/verify)

Your own site

<a href="https://agentmods.dev/skills/inflexa-ai/inflexa/verify"><img src="https://agentmods.dev/badge/skills/inflexa-ai/inflexa/verify/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/inflexa-ai/inflexa/verify"><img src="https://agentmods.dev/badge/skills/inflexa-ai/inflexa/verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 644 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00057	$0.00644
Opus 5.5	$0.00023	$0.00258
Sonnet 5.5	$0.00011	$0.00129
Haiku 4.5	$0.00006	$0.00064

Measured 1mo ago against content hash 12eead8cfc83, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

verify 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 1mo 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.

harness/.claude/skills/verify/SKILL.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Verifying @inflexa-ai/harness at the package boundary

The harness is a library: its surface is the npm package export, not a CLI or server. The embedding CLI consumes the published harness, so an uncommitted harness change is only observable by importing the freshly built package.

Recipe that works

Build: cd harness && npm run build (tsc → dist/), then bun link.

Postgres (this machine uses podman, not Docker Desktop):

podman run -d --name verify-pg -e POSTGRES_USER=cortex -e POSTGRES_PASSWORD=dev \
  -e POSTGRES_DB=cortex -p 127.0.0.1:5599:5432 pgvector/pgvector:pg18
podman exec verify-pg pg_isready -U cortex   # poll until ready (~3s)

Scratch consumer in the session scratchpad: package.json with "type": "module" and deps neverthrow + zod (match harness majors), then bun link @inflexa-ai/harness && bun install. Run scripts with node (harness runtime is Node; bun is test-only).
Import ONLY from @inflexa-ai/harness (the barrel). Useful exports: createPool, initCortexState(pool), runAgent, defineTool, passthroughStep, makeLocalAuth, plus whatever seam is under test.

Gotchas

createPool takes port as a string and requires sslMode: "disable" for a local container — default config path expects SSL.
A fake AgentChat must return neverthrow okAsync({message, finishReason}) and SHOULD honor the signal argument (if (signal?.aborted) throw signal.reason) — the real AI SDK provider does, and abort-path behavior is wrong without it (chat wires no isFatalLoopError; cancellation exits via the provider call).
Session shape: {identity:{user}, scope:{kind:"analysis", analysisId, threadId}, provenance:{agentId, callPath:[agentId]}, auth: makeLocalAuth()}.
Spawned helper processes exit fast — guard child.exitCode !== null before awaiting its exit event, or the await never settles (Node exit code 13).
TRUNCATE cortex_* tables between runs for idempotent assertions.
Cross-process claims (polling gateways, ledgers) deserve a real second process: spawn a sibling node script with its own pool.

Read the full file on GitHub · 45 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.

1mo ago First seen · 45 lines · 57 tokens per session scan A 12eead8cfc83

Subscribe to this mod's changes

verify is a skill published in the GitHub repository inflexa-ai/inflexa (35 stars, last pushed today), licensed Apache-2.0. It adds 57 tokens to every session and 644 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-09.

Databases PostgreSQL Testing agentic-ai agentic-orchestration bioinformatics computational-biology drug-discovery translational-medicine

test-database-seeding

Use when a test needs a database - spin up a dedicated, disposable test database (Testcontainers) seeded with known data, never a shared or production database.

makifbaysal/tasktrooper · 37 tokens

postgresql-table-design

Use this skill when designing or reviewing a PostgreSQL-specific schema. Covers best-practices, data types, indexing, constraints, performance patterns, and advanced features.

wshobson/agents · 37 tokens

prisma-postgres-setup

Set up a new Prisma Postgres database and connect it to a local project using the Management API. Use when asked to "set up a database", "create a Prisma Postgres project", "get a connection string", "connect my app to Prisma Postgres", or "provision a database".

nitrocloudofficial/nitrostack · 67 tokens

prisma-postgres

Prisma Postgres setup and operations guidance across Console, create-db CLI, Management API, and Management API SDK. Use when creating Prisma Postgres databases, working in Prisma Console, provisioning with create-db/create-pg/create-postgres, or integrating programmatic provisioning with service tokens or OAuth.

nitrocloudofficial/nitrostack · 63 tokens

moai-platform-database-cloud

Cloud database platform specialist covering Neon (serverless PostgreSQL), Supabase (PostgreSQL 16 with real-time), and Firebase Firestore (NoSQL with offline sync). Use when choosing or setting up cloud databases.

modu-ai/moai-adk · 51 tokens

Drizzle ORM Testing

Testing patterns for Drizzle ORM covering migration testing, query builder testing, transaction testing, and database integration testing with PostgreSQL, SQLite, and MySQL.

PramodDutta/qaskills · 36 tokens