How to use the Litigation Chronology Builder agent skill

What the skill makes the agent do

Once the skill is installed, you can ask the agent to build a chronology from a production, a matter folder or files you upload. The agent extracts dated events, merges duplicates, and lists each event with its sources.

Each event is tagged red, yellow or grey for how important it is to the case theory. Red marks events that matter most, yellow marks supporting events that are open to challenge, and grey marks background. Where a source could not be read, the output notes it under Gaps so you know what is missing.

The skill can also produce variants for a statement of facts or for a single witness. The output is saved as a chronology.md file in the matter's folder.

• Dated event list with duplicates merged
• Significance tags: 🔴 high, 🟡 contested or impeachable, ⚪ background
• Gaps section for sources the agent could not read
• Variants: working chronology, statement of facts (sof), or witness-specific timeline

Install the skill from the ZIP download

Download the ZIP for the skill from this page and unzip it. The skill's main file is SKILL.md. You can install it in one of two places, depending on whether you want it available everywhere or only in one project.

For all of your projects, unzip it into your user skills folder. For a single project, unzip it into that project's .claude/skills/ folder.

If you use the Claude apps instead of Claude Code, open Settings, go to Capabilities, then Skills, and upload the skill there.

• All projects: unzip into ~/.claude/skills/
• One project: unzip into the project's .claude/skills/ folder
• Claude apps: Settings → Capabilities → Skills → upload

Set up what the skill needs before the first run

The skill reads your matter setup from a plugin configuration folder. That folder must contain matter.md, CLAUDE.md and a matter log called _log.yaml. The matter intake step creates these files, so run it for each new matter first.

The skill refuses to build a chronology for a matter that is not in the matter log. Running the intake step performs the conflicts check and writes the matter's row to _log.yaml.

If you want the agent to read your document systems, you need access to the declared sources, such as Google Drive, SharePoint, Gmail, iManage or an eDiscovery platform. Connectors for Everlaw, Relativity, DISCO or Aurora are optional.

• Claude Code, or another compatible SKILL.md agent
• Configuration folder at ~/.claude/plugins/config/claude-for-legal/litigation-legal/ containing matter.md, CLAUDE.md and _log.yaml
• Matter must be intaken first: run /litigation-legal:matter-intake
• Access to declared document sources if you want them read; eDiscovery connectors are optional

Trigger the skill with natural requests

You do not need to remember a command. The skill runs when you ask for a chronology or timeline from a production or matter file. The source lists phrases such as 'chron from the production' and 'what happened when' as triggers, and also covers requests for a working, statement-of-facts or witness-specific timeline.

You can also invoke it directly with the argument hint [slug] and an optional format flag: working, sof, or witness followed by the witness name. Use --matter for matter-centric work and --documents for work on a document set.

• 'Build a chronology from the production for the Acme matter.'
• 'Chron from the production, statement of facts format.'
• 'What happened when, based on the matter folder?'
• 'Make a witness-specific timeline for the depo prep.'
• Direct form: /chronology [slug] --format=sof

Understand the two modes

The skill has two modes. --matter is the default for in-house litigation counsel. It reads the case theory and key facts from matter.md and pulls from your declared document-storage sources. It treats history.md as an internal log and keeps it out of the chronology.

--documents is the default for associates and paralegals. It works from an eDiscovery export, a custodial file set or a Bates-numbered production, and cites Bates numbers. It does not require a matter slug, but its output should be treated as pre-matter research rather than filed as matter work product.

If your role is solo or other, the skill defaults to --matter and mentions both modes on the first run so you can choose.

• --matter: matter-centric, uses matter.md, the conflicts gate applies
• --documents: production-centric, Bates citations, no matter slug required

Check the output before you rely on it

The output is AI extraction. It can miscall significance and it can miss events. Review every red-tagged entry before you use the chronology. The skill asks you to do this at the end of each run: 'Here's what I built. Scan the 🔴 entries — anything I miscalled?'

If a prior chronology.md exists for the matter, the version number increases and the skill shows a summary of the differences. The significance tags follow the side you are working for: plaintiff or defense. If you have not said which side, the skill asks.

• Scan every 🔴 entry against the source documents
• Check the Gaps list for sources that could not be read
• Confirm the side framing at the top of the output is the one you want
• Compare against the prior version when one exists

Safety notes: disclosed documents, privilege and network use

Before starting, the skill asks whether any documents were obtained through disclosure or discovery. Such documents can carry use restrictions. For example, under England and Wales CPR 31.22, disclosed documents may be used only for the proceedings in which they were disclosed, unless the court permits otherwise, the disclosing party consents, or the document has been read in open court. US protective orders and Rule 26(c) can impose similar limits. If you cannot confirm the use is permitted, the skill flags it and asks you to confirm before it continues.

The skill also runs a privilege check first. Matter files and productions can contain privileged material, and extracting it into a chronology that is later shared can risk waiver.

The skill does not run scripts. It does use the network, because it reads the document sources you declare, such as Google Drive, SharePoint, Gmail or iManage.

• Do not use disclosed documents for other cases or commercial purposes without permission, which can breach implied undertakings or protective orders
• Confirm the use before the skill proceeds when documents came through disclosure
• Runs scripts: no
• Uses network: yes, to read declared sources