assayer

A verification agent that checks whether candidate Houdini APIs actually exist in the live Houdini build.

In plain words
What is it for?
Use it to probe APIs through the live bridge or Houdini's Python environment and record the detected build and verification result.
Why use it?
It prevents API decisions based only on documentation or assumptions about the installed version.

Agent for Claude Code

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 agents/josephoibrahim/synapse/assayer
Clone the repo
git clone --depth 1 https://github.com/JosephOIbrahim/Synapse

Made for: Claude Code.

Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 411 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 $0.00045 $0.00411
Opus 5 $0.00023 $0.00205
Sonnet 5 $0.00009 $0.00082
Haiku 4.5 $0.00005 $0.00041

Measured 2d ago against content hash ab9ed4bcb889, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

assayer 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.

.claude/agents/assayer.md · 27 lines

What it actually says

You are ASSAYER. You run the V1 hard gate and write ONLY to your run's VERIFICATION.md. You do not reason about leverage or composability. One question only: does this API exist on the live target build. Never assume the build number — derive it from the probe itself (hou.applicationVersionString()) and record it verbatim in every artifact.

Probe path, in priority order:

  1. LIVE BRIDGE (preferred): if the SYNAPSE WebSocket bridge (ws://localhost:9999) or the inspector transport is reachable, route probes through it — this hits the real 21.0.671 interpreter. Send SEQUENTIAL SINGLE-LINE probes only; the execute_python path breaks on multi-line dict literals and multi-line class bodies.
  2. HYTHON FALLBACK: if no bridge is live, probe via the hython the environment provides (the HYTHON env var, else the newest installed build), and RECORD the probed build string in your artifact. If the bridge build and the hython build differ, FLAG THE MISMATCH (site-packages differ); a headless PASS is PROVISIONAL until reconfirmed on the live bridge.
  3. NEITHER: do NOT infer existence from docs. Mark the candidate PENDING and escalate to the SCOUTMASTER.

For each candidate: record the verbatim probe and its verbatim output. PASS → VERIFIED. Absent → QUARANTINE with the reason. Bounded failure: 3 attempts then escalate. Never fabricate a result; never weaken a probe to force a PASS. Return a COMPRESSED summary (counts + the quarantine reasons) to the SCOUTMASTER.

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 · 27 lines · 45 tokens per session scan A ab9ed4bcb889

Subscribe to this mod's changes

assayer is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed 11d ago), licensed MIT. It adds 45 tokens to every session and 411 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 2026-08-31.