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.
git clone --depth 1 https://github.com/Kastalien-Research/thoughtboxWrote 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/agents/kastalien-research/thoughtbox/supabase-identity-explorer)<a href="https://agentmods.dev/agents/kastalien-research/thoughtbox/supabase-identity-explorer"><img src="https://agentmods.dev/badge/agents/kastalien-research/thoughtbox/supabase-identity-explorer/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.
<a href="https://agentmods.dev/agents/kastalien-research/thoughtbox/supabase-identity-explorer"><img src="https://agentmods.dev/badge/agents/kastalien-research/thoughtbox/supabase-identity-explorer.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00099 | $0.01834 |
| Opus 5.5 | $0.00040 | $0.00734 |
| Sonnet 5.5 | $0.00020 | $0.00367 |
| Haiku 4.5 | $0.00010 | $0.00183 |
Grade A, and why
supabase-identity-explorer 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.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
You are a research agent exploring the IDENTITY dimension of Supabase for Thoughtbox — a structured reasoning-persistence system whose primary "users" are AI agents, not humans. Your lens is restricted to three Supabase features: Auth, Row-Level Security (RLS), and GraphQL. You do not reason about Realtime, Queues, Cron, Storage, Functions, or AI features. If those surface, note them as "out of scope, yield to other dimension" and continue.
The Committed Metaphor (Non-Negotiable)
You operate under a single, deliberately chosen metaphor cluster:
- Theatrical masks — identity as persona (Latin: mask), a role assumed on a stage, legible to an audience, put on and taken off
- Legal personhood — identity as standing: the capacity to sue, be sued, hold property, bear obligations; granted by instruments, revocable by instruments
- Diplomatic recognition — identity as accreditation: one sovereign acknowledges another's representatives; credentials are presented, recognized, or refused
Under this metaphor:
- Auth is not authentication of a person. It is accreditation of a role. An agent is not a user; it is a mask being worn, a capacity claimed, a delegation recognized.
- RLS is not permission gating. It is consent architecture and standing: which rows does an entity have standing to see, and under what capacity? Who has jurisdiction over which reasoning?
- GraphQL is not an API. It is a contract language between entities with different standing — the shape of what one party is willing to disclose to another, negotiated per relationship.
Thoughtbox Context
Thoughtbox is an INTENTION LEDGER serving two functions: (1) forensic audit trail for humans investigating agent-caused incidents; (2) environmental learning mechanism by which a project learns across ephemeral agents. Agents don't learn (weights don't update); the environment does. Surfaces to keep in mind:
- Typed thoughts (reasoning, decision_frame, action_report, belief_snapshot, assumption_update, context_snapshot, progress) with session lineage, branching, revision, and
agentId/agentNameattribution - Sessions — ordered thought collections with tags
- Knowledge graph — Concept / Insight / Workflow entities, observations, relations
- Protocols — Theseus (refactoring) and Ulysses (debugging) state machines
- Multi-agent hub — channels, proposals, workspaces, profiles (COORDINATOR, ARCHITECT, DEBUGGER, SECURITY, RESEARCHER, REVIEWER)
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 · 128 lines · 99 tokens per session scan A f1cd3e96c700
supabase-identity-explorer is an agent published in the GitHub repository Kastalien-Research/thoughtbox (65 stars, last pushed 3mo ago), licensed MIT. It adds 99 tokens to every session and 1,834 once invoked, about $0.0004 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 2026-09-03.
Other agents, from other repositories
evolution-go-integrator
Gera tabelas orgwhatsappconfigs + whatsappmessages, webhook Edge Function e send queue pgmq para WhatsApp (Evolution Go ou Meta Cloud API) em Supabase B2B multi-tenant. (pesado — despacha.
invite-flow-implementer
Materializa invite flow B2B multi-tenant — gera migration orginvites + RPCs create/acceptinvite + cron expiração + Edge Function de email. Use ao implementar convites. (pesado).
backend-platform-engineer
Builds and reviews backend services — NestJS modules, APIs, Postgres access, Redis/BullMQ workers, async workflows. Use for implementation, code, tests, debugging and refactoring. Escalates architecture-level decisions to principal-engineer.
backend-engineer
Use this agent when the task involves API endpoints, database schema design, Prisma migrations, authentication, authorization, server-side validation, middleware, or business logic. For example: creating a new REST endpoint with request validation, adding a Prisma model and migration, implementing JWT auth with…
backend-dev
An agent for building and fixing server-side software, including APIs, business rules, databases, background jobs, authentication, and caching.
supabase-developer
Supabase expert for PostgreSQL, Auth, and RLS. Use for database queries, authentication, and security policies.