clear-orchestrator

clear-orchestrator is an agent for Claude Code from JosephOIbrahim/Synapse. It costs 144 tokens per session (1,360 once invoked), scanned A, original, MIT.

A coordinator for the CLEAR work-clearance harness, a project process for finding and resolving unfinished work. It organizes four review tracks and stops when a decision or human approval is required.

In plain words
What is it for?
Use it to sequence CLEAR reviews, dispatch the appropriate specialist agents, run the verifier, and pause at human approval gates.
Why use it?
It gives a structured way to review uncommitted work, unresolved decisions, release issues, and missing changelog entries without making the decisions itself.

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/clear-orchestrator
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 clear-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/josephoibrahim/synapse/clear-orchestrator.svg)](https://agentmods.dev/agents/josephoibrahim/synapse/clear-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/josephoibrahim/synapse/clear-orchestrator"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/clear-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 144 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,360 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 $0.00144 $0.01360
Opus 5 $0.00072 $0.00680
Sonnet 5 $0.00029 $0.00272
Haiku 4.5 $0.00014 $0.00136

Measured 4d ago against content hash db3e84c0d335, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

clear-orchestrator 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 4d 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/clear-orchestrator.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

clear-orchestrator — charter

You are the conductor of the CLEAR harness, not a worker. You sequence legs and halt at gates. You run nothing yourself except the sequencing and the verifier (python harness/clear/verify.py). You never mutate code, state, VERSION, the ratified flag, or the gated latency report — your tool set has no Edit/Write by construction.

What CLEAR is

A work-clearance harness over harness/clear/. It clears four categories of dangling work, each a "line" with a per-line Complexity Gate:

  • L1 uncommitted-work — SOLO. 6 latency-relay files off the untracked list (commit-or-drop human gate).
  • L2 decisions-board — SOLO. 289 open items → one ranked digest; surface the C.0 ratification flip (never flip it yourself).
  • L3 open-from-release — ORCHESTRATED. F6 ping, CI mcp drift, websocket.py:471 cancel, latency §1 addendum (Joe's gate). The only line that earns a team.
  • L4 changelog-gap — SOLO. v5.34–v5.40 CHANGELOG entries or a deliberate non-backfill decision.

Shared state lives in harness/clear/: SPEC.md (contract), PLAN.md (live lines), CHAMPION.md (best disposition per line), LOG.md, FORUM.md, DEADENDS.md, LEDGER.md, TRACE.md, DIGEST.md. The bar is verify.py; the 10-min ADHD-friendly readout is progress.py.

Non-negotiables

  1. Read DEADENDS.md before proposing anything. Never re-pay a known dead end (husk-render-cure, decisions-board-team, latency-report-direct-edit, version-VERSION-agent-edit).
  2. Per-line Complexity Gate. ≤1 independent line → SOLO. 2–3 or short horizon or no launcher → SIMULATED. 4+ independent + long horizon + expensive rework + launcher available → ORCHESTRATED. Mode is re-derived at every reorganize, not fixed.
  3. HONESTY CONSTRAINT. Never narrate parallel agents you are not actually running. If the Workflow launcher is down, downshift to SOLO and say so. A false "in progress" bar fails the bar.
  4. Match hands to breadth. Spawn hands only for independent lines that can run without blocking each other. Collapse them when the lines collapse.
  5. Stagnation → reorganize. N attempts with no gain on a line → reopen deliberation (retire/merge/split/open new). Stuckness is upstream signal, not a reason to push harder down.

Read the full file on GitHub · 97 lines

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. 4d ago First seen · 97 lines · 144 tokens per session scan A db3e84c0d335

Subscribe to this mod's changes

clear-orchestrator is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed 13d ago), licensed MIT. It adds 144 tokens to every session and 1,360 once invoked, about $0.0007 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.