record-cassette

A recording step that runs a probe against a real external service and saves the request and response as a cleaned test fixture called a cassette. A cassette lets later tests replay the service interaction without calling the service again.

In plain words
What is it for?
Use it after writing a probe to capture HTTP APIs or other external services, scrub sensitive and unstable data, and document the confirmed interaction.
Why use it?
It makes external-service tests repeatable and safe to commit by removing secrets and changing response details. It also records the exact interaction shape needed for implementation.

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

Made for: Claude Code, Codex.

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 362 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 $0.00028 $0.00362
Opus 5 $0.00014 $0.00181
Sonnet 5 $0.00006 $0.00072
Haiku 4.5 $0.00003 $0.00036

Measured 2d ago against content hash d01306a58f21, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

record-cassette 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/record-cassette/SKILL.md · 15 lines

What it actually says

Record Cassette

  1. Load [[software-craft/external-fixtures]] — the recorder and scrub-field kind-dispatch table.
  2. Execute the probe script and capture the real request/response with the transport-appropriate recorder.
  3. IF HTTP THEN record with vcrpy and decode_compressed_response=True so gzip/br bodies are stored decoded, not opaque blobs per [[software-craft/external-fixtures]]. Dispatch the recorder on the service kind (HTTP API, external database, message queue, object storage, other) — the recorder is the one knob that varies per kind.
  4. Scrub the cassette to deterministic and safe-to-commit. IF HTTP THEN filter_headers strips Authorization and before_record_response strips volatile headers — date, age, server, last-modified, content-encoding, transfer-encoding, vary, cf-ray, report-to, nel. Dispatch the scrub-field list on the service kind.
  5. IF a cassette has an opaque/churning body or a leaked secret THEN it fails the external contract — re-scrub before commit.
  6. Append to the shared external-contracts notes the confirmed endpoint base, auth scheme, gotchas, AND the exact request shape captured per interaction (method, path, query parameters, required headers). The cassette is the contract and vcr matches on the request, so the implementer must reproduce this exact shape — record it explicitly, do not leave it implicit in the cassette.
  7. IF the live exchange contradicts the research or the interview 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 · 28 tokens per session scan A d01306a58f21

Subscribe to this mod's changes

record-cassette is a skill published in the GitHub repository nullhack/temple8 (11 stars, last pushed 27d ago), licensed MIT. It adds 28 tokens to every session and 362 once invoked, about $0.0001 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.