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.
npx agentmods add skills/jinghao67/conductor/conductor-ccnpx skills add Jinghao67/conductor --skill conductor-ccgit clone --depth 1 https://github.com/Jinghao67/conductorWrote 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/skills/jinghao67/conductor/conductor-cc)<a href="https://agentmods.dev/skills/jinghao67/conductor/conductor-cc"><img src="https://agentmods.dev/badge/skills/jinghao67/conductor/conductor-cc.svg" alt="Measured on agentmods" 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 | $0.00089 | $0.01000 |
| Opus 5 | $0.00044 | $0.00500 |
| Sonnet 5 | $0.00018 | $0.00200 |
| Haiku 4.5 | $0.00009 | $0.00100 |
Grade A, and why
conductor-cc 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.
How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conductor (Claude Code)
A context-isolation + branch-registry protocol for Claude Code. One master session stays clean; messy work happens in separate branch sessions and throwaway subagents. Unlike a plain skill, the hard rules here are enforced by hooks, not by your goodwill — so this skill only has to carry the judgment the hooks can't.
The one principle
The master holds only distilled knowledge. All raw content is quarantined.
| Raw content | Quarantine | Master gets |
|---|---|---|
| a branch's messy history | a separate user-driven session | a completion report |
| codebase source | the Explore subagent | conclusion + file:line anchors |
| paper/web content | the cd-research subagent | distilled notes + sources |
What the harness enforces for you (don't re-litigate it)
- SessionStart hook injects role context: master → snapshot + Today View; a bound branch → only its brief. A branch literally cannot see master/sibling history.
- PreToolUse hook = three gates: (a) read-firewall — broad/external reads in master are denied, forcing you to delegate to Explore/cd-research; (b) merge-gate — the master snapshot can't absorb a branch without an approved report; (c) branch-write-guard — a branch can't write master records.
- PostToolUse hook keeps
registry.json→branch-map.mdin sync and performs privileged state transitions.
When a read is denied in the master, that is by design: spawn an Explore subagent for codebase questions, or cd-research for web/papers, and absorb only the returned conclusion. Do not fight the firewall; /cd-status --unlock exists if the user truly needs it off.
What YOU must judge (the soft layer)
- Routing — where does a request belong?
- master: changes to goals, scope, constraints, the branch map, merges, snapshots.
- a branch: exploratory / implementation / research / review work with many intermediate details.
- explainer sidecar: long explanations, tutorials, "I don't fully get this yet."
- Dependency / wave planning — never assume all branches run in parallel. Classify each:
ready_parallel(runs now from the current snapshot),dependent(needs another branch's output),gate(a review/merge/decision),optional,explainer. Only open current-wave branches. Keep ≤2 active interactive branches; suggest parking/archiving before a third. - What to merge — at
/cd-merge, compress to a short approved record, normalize terminology, and separate proposed global decisions for explicit user confirmation. Reject reports that don't meet the brief (the cd-merge-verify subagent backs you up). Explainer output never merges by default. - Card before session — create a branch card + brief (
/cd-branch) before anyone opens a session. If dependencies are ambiguous, ask the user before creating threads. - Staleness — if the master snapshot advances in a way that invalidates an active brief, mark it stale and offer to refresh; don't silently sync.
- Global decisions — a decision made inside a branch is not binding until confirmed in the master.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 54 lines · 89 tokens per session scan A b2434d7a2d80
conductor-cc is a skill published in the GitHub repository Jinghao67/conductor (106 stars, last pushed 2mo ago), licensed MIT. It adds 89 tokens to every session and 1,000 once invoked, about $0.0004 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…