defer

defer is a skill for Claude Code, Codex from fledgeling-co/fledgeling-plugins. It costs 254 tokens per session (4,006 once invoked), scanned A, original, MIT.

A model-routing skill that chooses a model, effort level, and verification method for delegated work. It uses task class, work shape, model capability, and measured remaining capacity to produce one command.

In plain words
What is it for?
Use it to route implementation and other task types, generate the exact command for the selected lane, and verify the resulting run.
Why use it?
It puts routing decisions in one policy so different skills do not choose models inconsistently or claim success without checking that the route ran.

Skill for Claude CodeCodex

Written for Claude Code and Codex: ${CLAUDE_PLUGIN_ROOT} variable, but also runs codex exec. Also seen: positional $N argument; mentions Claude Code; mentions Codex.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the defer plugin — 1 skill shipped together

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add fledgeling-co/fledgeling-plugins
Claude Code
/plugin install defer

Made for: Claude Code, Codex.

Or install defer, the plugin that ships this one along with the rest of its 1 skill.

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 defer

README.md
[![agentmods](https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/defer.svg)](https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/defer)
Your own site
<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/defer"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/defer.svg" alt="Measured on agentmods" height="20"></a>
Per session 254 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,006 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 $0.00254 $0.04006
Opus 5 $0.00127 $0.02003
Sonnet 5 $0.00051 $0.00801
Haiku 4.5 $0.00025 $0.00401

Measured 5d ago against content hash 1bf49fdd8876, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

defer 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.

The scan reads SKILL.md. This mod also ships 6 executable files (scripts/gen_capability.py, scripts/lane_pick.py, scripts/lane_probe.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

plugins/defer/skills/defer/SKILL.md · 300 lines

How it starts

The opening of the file, as written. The whole thing — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.

defer

Handing work to another model is three decisions, and skills that make them inline get them subtly wrong: which model, at what effort, and how you know it really ran. This skill holds all three in one place so that every skill routes the same way, and so that changing the policy is one edit rather than fourteen.

The routing rule is not "use the best model". It is: the work class decides the family, the work shape decides which lanes are good enough, and measured headroom decides which of those runs it.

Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. It makes running lane_pick.py the only source of a route, turns the three rules above the class table into a bound ledger read back off the emitted argv, and requires the lane's receipt rather than its flags. Other models skip it.

Route

Run this. It reads the policy and the meters and prints a command:

python3 ${CLAUDE_PLUGIN_ROOT}/skills/defer/scripts/lane_pick.py --task <class> [--shape <shape>]

The classes, and what each one is for:

--task For Lanes Effort
implementation writing code shape decides, then headroom pinned per lane
completeness the critic that finds what was promised and not delivered glm · grok · gemini high · xhigh · high
general anything that is neither a referred decision nor a verdict gpt-5.6-terra high
referral a decision or fork put to another model gpt-5.6-sol · claude-fable-5 medium · high
verification grading delivered work; same-family validation claude-opus-5 xhigh
design-review judging rendered UI claude-opus-5 · claude-fable-5 xhigh · high

--json gives the same answer as a structured object with the argv and env ready to spawn. --report prints every lane's meter without choosing, and --matrix prints the measured capability table without choosing either.

Read the full file on GitHub · 300 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. 5d ago First seen · 300 lines · 254 tokens per session scan A 1bf49fdd8876

Subscribe to this mod's changes

defer is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed today), licensed MIT. It adds 254 tokens to every session and 4,006 once invoked, about $0.0013 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

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…

microsoft/ai-agents-for-beginners · 200 tokens

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…

microsoft/vscode · 72 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

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…

vercel/next.js · 95 tokens