Guide · Agent skill
How to use the Portfolio Status Rollup agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The portfolio-status skill is a reusable instruction set for Claude that reads a local log of your legal matters and produces a single status report: how many matters are open, their risk levels, upcoming deadlines, stale matters, exposure totals and flagged anomalies. It is written for in-house or litigation counsel who need an overview of many active matters before a meeting. After this guide you will know what the skill needs, how to install it, how to ask for a rollup in plain language, and what to check in the output.
Portfolio Status Rollupanthropics/claude-for-legal · Apache-2.0 Download the skill Quick start
- 1Download the ZIP. Download the portfolio-status skill ZIP from this page.
- 2Unzip the skill folder. Extract it so SKILL.md sits in ~/.claude/skills/portfolio-status/ or a project's .claude/skills/portfolio-status/ folder.
- 3Or upload it in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the ZIP.
- 4Check your matter log exists. Confirm _log.yaml is present and maintained in the litigation-legal matters config folder.
- 5Check the risk calibration file. Confirm the litigation-legal CLAUDE.md file is in the same config folder.
- 6Ask for a rollup. Say something like "where do we stand?" or run /portfolio-status with a flag such as --risk=high.
- 7Review the anomalies. Work through the anomaly list first and decide what needs action; the skill does not decide for you.
What an agent skill is and what this one does
An agent skill is a folder containing a SKILL.md file of instructions that Claude loads when a request matches the skill's description. This skill is part of the anthropics/claude-for-legal project, in the litigation-legal folder, and its file is SKILL.md.
When it runs, the agent reads your matter log and your risk calibration file, then produces a rollup. The report covers risk distribution, deadlines in the next 14, 30 and 60 days, matters with no update in more than 30 days, materiality totals, stage distribution, and anomalies. The skill surfaces items for you to decide on; it does not make decisions itself.
- Risk distribution by Critical, High, Medium and Low
- Upcoming deadlines grouped by 14, 30 and 60 day windows
- Stale matters with no update in more than 30 days
- Materiality totals (Reserved, Disclosed, Monitored, None)
- Stage distribution across the matter lifecycle
- Anomalies such as overdue deadlines, unresolved conflict checks and high-risk matters without outside counsel
Install the portfolio-status skill
Download the ZIP for this skill from this page. Unzip it so that the portfolio-status folder, containing SKILL.md, sits in one of these locations: ~/.claude/skills/ for all your projects, or .claude/skills/ inside a specific project folder.
You can also install it through the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the ZIP there.
- Personal install: place the folder in ~/.claude/skills/
- Project install: place the folder in the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Set up the files the skill reads
The skill needs two files in your Claude config folder. The first is your matter log, which must be maintained and kept up to date. The second is the risk calibration file, which tells the agent how to interpret the risk and materiality fields.
Without the matter log the skill has nothing to read and will not work. The skill does not create this file for you.
- ~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/_log.yaml (the matter log, source of truth)
- ~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md (risk calibration)
~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/_log.yamlAsk for a rollup in natural language
You do not need a special command. The skill is triggered when you ask questions such as where you stand, how many open matters you have, or for a status across all active matters. The skill's description gives these as examples of matching requests.
The skill also documents optional flags that narrow the report. Flags are listed in its argument hint as --all, --risk=high and --stale, and the source also lists --risk for critical, medium and low, --type=employment and --owner=[name].
- "Where do we stand on our matters?"
- "How many open matters do we have?"
- "Give me a portfolio rollup across all active matters."
- "Show me the high-risk matters." (maps to --risk=high)
- "Which matters have gone stale?" (maps to --stale)
/portfolio-status --risk=highRead the report and the anomaly list
The report starts with active matter counts, then a risk table, deadlines, materiality, and stage. Closed matters are excluded by default, and they appear only when you use --all.
The anomalies section is the most useful part. It lists overdue deadlines, stale matters, unresolved and bypassed conflict checks, high or critical matters without outside counsel, reserves that have not been updated in over 60 days, holds not issued on active litigation, and missing required fields. The closing advice names what to look at first, if anything stands out.
Exposure figures are rough midpoints, and the skill labels them that way. Treat them as orientation, not as numbers for a reserve or a board paper.
- Overdue deadlines: next deadline has passed and the matter is not closed
- Stale: no update in more than 30 days
- Conflicts unresolved: status is pending or not-run
- Conflicts override active: stays flagged until manually cleared
- High or critical risk with no outside counsel assigned
- Hold gap: active or threatened matter without a legal hold issued
Safety, privacy and limits
According to the skill's description, it runs no scripts and makes no network calls. It reads local files only. The report summarizes matter information that may be privileged or sensitive, and the files live in your local Claude config folder, so run it only on a machine you control and that is secured.
The skill is a working-memory rollup. It is not a system of record, and a lawyer should check its outputs before acting on them.
- Runs scripts: no
- Uses network: no
- Run on a secured machine when matter data is confidential
- Verify flagged items against your own matter records
What people use it for
- A litigation counsel checking which matters have deadlines in the next 14 days before a weekly meeting.
- An in-house lawyer reviewing high-risk matters that have no outside counsel assigned.
- Legal operations staff producing a quick risk and deadline overview across a matter portfolio.
- A team spotting matters that have gone quiet for more than 30 days and need an update.
- Reviewing reserved matters whose last update is more than 60 days old.
Frequently asked questions
Is the portfolio-status skill free to use?
The source does not state a price or licence terms beyond the listed licence, Apache-2.0. Check the repository's licence file for the terms that apply to you.
What do I need to run the portfolio-status skill?
You need a maintained _log.yaml matter file in the litigation-legal matters config folder, the litigation-legal CLAUDE.md risk calibration file in the same config folder, and a Claude environment that can read local files. Without the matter log the skill will not work.
Where do I put the portfolio-status skill files?
Unzip the folder into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings, Capabilities, then Skills.
Does the skill send my matter data over the internet?
The listed metadata says it uses no network and runs no scripts. It reads the local files you point it to, but the report contains sensitive matter information, so keep the machine secure.
Does it replace a matter management system?
No. The skill describes itself as a working-memory rollup, not a system of record. Use your matter management system as the source of record and treat this report as a quick overview.
Why does the skill show no results or fail?
The most likely cause is a missing or unmaintained _log.yaml file, or missing CLAUDE.md risk calibration file in the expected folder. Confirm both paths listed in the requirements exist before rerunning the skill.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you already keep a maintained _log.yaml matter file and want a fast, scannable overview of risk, deadlines and anomalies across many active matters. Pick a matter management system or another tool if you need a system of record, do not maintain a matter log, or cannot keep confidential matter data on a secured local machine. A lawyer should check every output before acting on it.
Download the skill