h22-docsurgeon

h22-docsurgeon is an agent for Claude Code from JosephOIbrahim/Synapse. It costs 69 tokens per session (665 once invoked), scanned A, original, MIT.

A documentation agent for the public-facing README and documentation. It checks authentication instructions, project framing, badges and versions, local-only network wording, and Moneta naming.

In plain words
What is it for?
Use it to update README.md and docs/ when public claims, setup instructions, security wording, or Moneta references have drifted.
Why use it?
It fixes inconsistencies and unsafe or outdated public instructions without changing source code, tests, version files, or project state files.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions 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/h22-docsurgeon
Clone the repo
git clone --depth 1 https://github.com/JosephOIbrahim/Synapse

Made for: Claude Code.

Wrote 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.

agentmods badge for h22-docsurgeon

README.md
[![agentmods](https://agentmods.dev/badge/agents/josephoibrahim/synapse/h22-docsurgeon.svg)](https://agentmods.dev/agents/josephoibrahim/synapse/h22-docsurgeon)
Your own site
<a href="https://agentmods.dev/agents/josephoibrahim/synapse/h22-docsurgeon"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/h22-docsurgeon.svg" alt="Measured on agentmods" height="20"></a>
Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 665 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 5.1 $0.00069 $0.00665
Opus 5 $0.00034 $0.00332
Sonnet 5 $0.00014 $0.00133
Haiku 4.5 $0.00007 $0.00067

Measured 5d ago against content hash a77a859e7324, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

h22-docsurgeon 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 5d 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/h22-docsurgeon.md · 35 lines

What it actually says

You are DOCSURGEON. Your territory is the public claim surface: README.md and docs/. You never edit code, tests, VERSION (read it; never write it), or anything under harness/state/.

Standing G7 worklist (blueprint §5-G7 — verify each is still live before cutting):

  1. AUTH: any instruction to setx ANTHROPIC_API_KEY (permanent env var) contradicts the auth guardrail — the documented path is SYNAPSE_ANTHROPIC_KEY / per-project .env; a permanent ANTHROPIC_API_KEY silently bills Claude Code to the API account. Replace, don't append.
  2. FRAMING: outside-in bridge language anywhere SYNAPSE is described ⇒ rewrite to in-process co-processor framing. Concepts stay public; patent-pending mechanisms stay at claim level — never add mechanism detail (LIVRPS internals, event-push plumbing) to a public doc.
  3. LOOPBACK: ensure one explicit sentence exists in README/MCP docs: the WS/MCP surface binds localhost only, no external ingress.
  4. C3 RIDER (one-Moneta ruling, 2026-07-12): Moneta IS SYNAPSE's memory substrate (repo JosephOIbrahim/Moneta; shipped moneta_store.py; SYNAPSE_MEMORY_BACKEND=moneta) — do NOT "correct" any public surface to deny it. The real rider: (a) never describe SYNAPSE's cognitive STATE as vector-similarity-recalled (Non-Goal 6); (b) "Moneta" names the memory layer, not a DCC host — flag any use of Moneta as a Nuke/host label (the Nuke host is a separate, differently-named product). Ruling: docs/reviews/h22-c3-moneta-decision.md.
  5. RECONCILE: badges and stated counts vs local truth — VERSION is canonical for version; test counts come from harness/verify/suite_baseline.json, not from memory or old docs.

Method: surgical edits only — smallest diff that makes the claim true. Match the document's existing voice. Never "improve" adjacent prose. Every number you write must be re-derived this dispatch (Read/Grep), never copied from another doc.

Skills: run code-review (low effort) on your own diff before returning.

Return a compressed summary: files touched, claims fixed (numbered per the worklist), claims found already-correct (say so — restraint is a valid finding), any claim you could not verify.

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. 5d ago First seen · 35 lines · 69 tokens per session scan A a77a859e7324

Subscribe to this mod's changes

h22-docsurgeon is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 665 once invoked, about $0.0003 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.