split-work SKILL.md skillA
split-work from abhiramdwivedi/claude-ops is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add abhiramdwivedi/claude-ops/split-work@git:20261007.498562bPins git:20261007.498562b in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- abhiramdwivedi/claude-ops
- Path
plugins/subagent-limits/skills/split-work/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261007.498562b, committed Oct 7, 2026
- Size
- 4,786 bytes
- Repository stars
- 3
- License
- MIT
How to size and split work across subagents so no single agent grows huge. Use before starting any subagent or teammate for multi-step work, when planning parallel builders, when a subagent's report starts with "CHECKPOINT:", or when the person asks whether agents are sized well or why a session burns tokens.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Split work across agents You decide how work is split. The subagent-limits plugin measures each subagent's context and enforces two limits: at the first it asks the agent to checkpoint, and past the second it refuses the agent's tools. Those limits are a backstop for a bad split. A good split never reaches them. ## Why size matters Every request an agent makes re-sends its whole context. An agent's cost therefore grows with the square of its length. In one measured session of 26 subagents, the typical agent started near 47K tokens of context and read about 200K more before its first edit. The largest peaked near 770K. The long ones did most of the spending. Two things cut that cost. The first is ending agents before they grow large. The second, which matters more, is making a fresh agent cheap to start, so that handing off costs little. ## Choose the shape 1. **One agent** fits a job with one goal that touches a few files and that you could describe in a short paragraph. 2. **Parallel agents** fit independent jobs that edit different files. Check the file lists before you start them. Two agents editing the same file will collide. …
Read the whole file at its exact version.
How to install
mdr add abhiramdwivedi/claude-ops/split-work@sha256:7c6896191dacb48cThe line at the top of this page pins the label git:20261007.498562b, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_6tkpjkb7wtbzdf7k)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 7, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (4786 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
abhiramdwivedi/claude-ops · 3 stars · license MIT · pushed 2026-11-02 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_6tkpjkb7wtbzdf7k GET https://markdownregistry.com/api/v1/resolve?ref=abhiramdwivedi/claude-ops/split-work GET https://markdownregistry.com/api/v1/blob/7c6896191dacb48c64ee49575adf67a4cbe30841ec9ba9a35ed85e321ba6e958
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.