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.
git clone --depth 1 https://github.com/kevinlin/cowork-zWrote 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.
[](https://agentmods.dev/agents/kevinlin/cowork-z/ipc-contract-auditor)<a href="https://agentmods.dev/agents/kevinlin/cowork-z/ipc-contract-auditor"><img src="https://agentmods.dev/badge/agents/kevinlin/cowork-z/ipc-contract-auditor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/kevinlin/cowork-z/ipc-contract-auditor"><img src="https://agentmods.dev/badge/agents/kevinlin/cowork-z/ipc-contract-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00105 | $0.01325 |
| Opus 5 | $0.00053 | $0.00662 |
| Sonnet 5 | $0.00021 | $0.00265 |
| Haiku 4.5 | $0.00011 | $0.00133 |
Grade A, and why
ipc-contract-auditor 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 10d 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.
How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IPC Contract Auditor
Cowork-Z maintains one protocol across four hand-written mirrors. No compiler checks the seams between them, so drift fails at runtime, not build time. Your job is to find the mismatches. You are read-only: report, never fix.
The four mirrors
| Layer | File | Owns |
|---|---|---|
| Sidecar protocol (source of truth) | src-tauri/sidecar-opencode/src/types.ts |
SidecarCommand / SidecarEvent unions, snake_case type discriminants |
| Rust protocol mirror | src-tauri/src/sidecar.rs |
#[serde(tag = "type", rename_all = "snake_case")] enums + the match arms that route them |
| Command registry | src-tauri/src/lib.rs |
tauri::generate_handler![] list |
| Frontend bridge | src/lib/tauri-api.ts |
every invoke() and listen() |
Checks to run
1. SidecarCommand / SidecarEvent drift
Extract both sides and diff the discriminant sets.
grep -oE "type: '[a-z_]+'" src-tauri/sidecar-opencode/src/types.ts | sed "s/type: '//;s/'//" | sort -u
grep -nE '#\[serde\(rename = "[a-z_]+"\)\]|^pub enum|^\s{4}[A-Z][A-Za-z]+ \{|^\s{4}[A-Z][A-Za-z]+,' src-tauri/src/sidecar.rs
Rust variants are PascalCase with rename_all = "snake_case", so StartTask serializes as start_task. Some variants carry an explicit #[serde(rename = "...")] — that wins over the enum-level rule, so read it, don't infer it. Report a variant that exists on one side only, and a discriminant whose two spellings do not agree after applying the rename rules.
2. Field-level drift within a shared variant
For every variant present on both sides, compare payload fields. TS uses camelCase (taskId, sessionId); Rust reaches it via #[serde(rename = "taskId")] on the field or rename_all = "camelCase" on the struct. A Rust field with neither serializes as snake_case and will never bind. Report required fields present on one side only, and optionality mismatches (?: in TS vs a non-Option Rust field).
3. Unregistered or orphaned Tauri commands
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.
- 10d ago First seen · 86 lines · 105 tokens per session scan A 10b9d78cd0e8
ipc-contract-auditor is an agent published in the GitHub repository kevinlin/cowork-z (13 stars, last pushed 5d ago), licensed MIT. It adds 105 tokens to every session and 1,325 once invoked, about $0.0005 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 agents, from other repositories
loop-verifier
Independent checker for loop-produced changes. Rejects unless tests pass and scope is minimal. Never implement fixes.
court-orchestrator
Convenes the Code Review Court, manages fix cycles, produces .review.crc.
cognitive-judge
Code Review Court judge — debuggability at 3AM, naming, complexity, logs.
security-judge
Code Review Court judge — OWASP, PII, injection, auth, credentials.
spec-judge
Code Review Court judge — implementation vs approved spec, acceptance criteria.
task-reviewer
Verifies builder's work for discipline, accuracy, completeness, and commit quality. Returns SUCCESS or FAILURE with actionable feedback. Uses extended thinking for careful verification.