research-provider

A research step for learning how a specific external service really works from its official documentation before writing integration tests or probe code.

In plain words
What is it for?
Use it to document an API or other provider’s connection details, success and failure responses, configuration needs, and relevant security or data concerns.
Why use it?
It reduces mistakes around authentication, request formats, errors, limits, pagination, repeated requests, and required secrets.

Skill for Claude CodeCodex

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 skills/nullhack/temple8/research-provider
Any agent
npx skills add nullhack/temple8 --skill research-provider
Clone the repo
git clone --depth 1 https://github.com/nullhack/temple8

Made for: Claude Code, Codex.

Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 348 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 $0.00031 $0.00348
Opus 5 $0.00015 $0.00174
Sonnet 5 $0.00006 $0.00070
Haiku 4.5 $0.00003 $0.00035

Measured 2d ago against content hash 75ae3338e308, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research-provider 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.

.opencode/skills/research-provider/SKILL.md · 15 lines

What it actually says

Research Provider

  1. Load [[software-craft/external-fixtures]] — the kind-dispatch table (docs-focus, recorder, scrub-fields, default specialist).
  2. Identify the exact provider — the specific service and version, not just its kind.
  3. Research online: websearch the provider and webfetch its official docs. Read for reference, authentication/credentials scheme, client/SDK, error catalog, rate limits, pagination, idempotency, known gotchas. Do not write probe code until the docs are read.
  4. Determine the access shape: endpoint base/URI, protocol, credentials source (a named variable the user must supply in ~/.secrets/<project>.env), request/response shape, and both success and failure modes. Per [[software-craft/secrets-and-config]] the agent names the variable and instructs the user where to set it; it never holds or prints the value.
  5. Recommend at most one advisory specialist for this provider. IF the provider raises auth/secrets/PII concerns THEN recommend security; IF schema/message-shape THEN data; IF user-facing flows THEN ux; IF project-specific semantics THEN domain.
  6. Dispatch on the service kind (HTTP API, external database, message queue/event stream, object storage, other) for the docs-focus and default-specialist per [[software-craft/external-fixtures]] — the kind table is the one knob that varies.
  7. IF the docs reveal the service behaves differently than the interview assumed THEN signal a mismatch.
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 · 15 lines · 31 tokens per session scan A 75ae3338e308

Subscribe to this mod's changes

research-provider is a skill published in the GitHub repository nullhack/temple8 (11 stars, last pushed 27d ago), licensed MIT. It adds 31 tokens to every session and 348 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 2026-08-30.