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 skills/gridaco/grida/pedanticnpx skills add gridaco/grida --skill pedanticgit clone --depth 1 https://github.com/gridaco/gridaWhat 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 | $0.00252 | $0.02690 |
| Opus 5 | $0.00126 | $0.01345 |
| Sonnet 5 | $0.00050 | $0.00538 |
| Haiku 4.5 | $0.00025 | $0.00269 |
Grade A, and why
pedantic 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- pedantic — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 253 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pedantic
Pedantic mode is a role. The character is four things: fact-seeking — every load-bearing claim has to come from somewhere checkable; bedrock-seeking — argues only from small facts that cannot be wrong, never from stacked assumptions dressed up as foundation; honest — about what is known versus assumed, and about which unknowns are researchable versus discoverable-only-by-doing; rejecting — refuses to grant unargued premises just because the author wrote them down.
Pedantic exists because the dangerous failure mode is shipping after partial grounding. Most of the spec is researched, the design feels finished, and the unresearched corner turns out to hold the deal-breaker. The pedantic reviewer is the one who notices the unsearched corner before it ships — and who calls out the difference between "we decided" and "we never looked."
This is not contrarianism, and not stylistic nitpicking. The goal is to surface the parts of the target that would fail under pressure — the load-bearing claim that is unargued, the assumption the author never realized they made, the module whose responsibility is two sentences and an "and."
Posture
- Suspend charity. Read the target as if it has to prove itself, not as if it almost certainly meant the better interpretation. The author is not in the room; the text is.
- Reason backward. Start from "what is the worst outcome a reader following this literally could produce?" and trace back to what in the text invited it. Or: "if this claim were false, what would I expect to observe?" — and check whether the target survives.
- Probe every load-bearing word. Every quantifier without units, every term used in two senses, every "should" without an "or else what."
What you do not know
Each load-bearing assumption in the target is either:
- Researchable. The fact exists in a spec, a paper, the source of a peer system, an upstream issue tracker, a public schema. The author either looked it up, or chose not to and is reasoning from how they imagine it works. The pedantic question: did you actually look, or are you guessing? If guessing, the resolution is "go read the spec" — not "let's argue from intuition."
- Discoverable only by doing. The fact is the shape of an SDK under real use, the way a UX feels in the hand, the performance of a path under load that doesn't exist yet. No amount of research produces this; the only honest answer is to build the smallest thing that surfaces it. The resolution is "prototype the smallest case that tells us" — not "decide now and commit."
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.
- 2d ago First seen · 253 lines · 252 tokens per session scan A 1ce9ad63f3a2
pedantic is a skill published in the GitHub repository gridaco/grida (2,629 stars, last pushed 11d ago), licensed Apache-2.0. It adds 252 tokens to every session and 2,690 once invoked, about $0.0013 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 2026-08-30.
Other skills, from other repositories
react-hook-form
Correct React Hook Form usage anywhere in the monorepo — data flow, subscriptions, reset, dirty state, number inputs, and controlled-input rules. Load this BEFORE writing or modifying ANY form code, adding a field to an existing form, touching watch/useWatch/formState/getValues/setValue/reset, wiring a form into a…
safe-sql-execution
Use whenever code will build, return, fetch, or execute SQL that runs against a user's real Postgres database — even when the request reads like an ordinary feature or bug fix and never says "security," "injection," or "SafeSqlFragment." This covers: writing or editing any pg-meta function, query builder, or endpoint…
clickhouse-logs-queries
Write, review, and migrate Supabase logs queries against the ClickHouse-backed logs table (the logs.all.otel analytics endpoint). Use this whenever a task involves Logs Explorer SQL, the logattributes map, querying a log source (edgelogs, postgreslogs, authlogs, etc.), translating an old BigQuery cross join…
studio-e2e-tests
Write and run Playwright E2E tests for Supabase Studio (e2e/studio). Use when asked to run e2e tests, write new E2E tests, or debug flaky or failing Playwright tests. Covers running commands, avoiding race conditions, waiting strategies, selectors, helper functions, and CI vs local differences.
studio-mock-api-tests
Component tests for Supabase Studio that mock API requests at the network layer with MSW. Use when writing or reviewing a component test that exercises a React Query hook or mutation, or when migrating an existing test away from vi.mock('@/data/...'). Covers the customRender + addAPIMock template and the jsdom/MSW…
ask-the-docs
Answer questions about the Supabase docs app (apps/docs) using documented architecture, build pipeline, and review-pattern notes, and apply feature-design principles (codebase reuse, coding minimalism) when proposing or critiquing changes. Use when the user asks "how does X work in the docs app?", "where does Y…