Guide · Agent skill
How to use the CTO Advisor agent skill: setup and examples
Updated · written with AI from the project's own documentation and checked against our catalogue data
The CTO Advisor is an agent skill: a packaged set of instructions, frameworks and helper scripts that an AI agent (such as Claude) loads when a task matches it. Once installed, the agent can reason through technical leadership questions such as technical debt, team scaling, technology evaluation and architecture decisions using structured workflows instead of generic advice. This guide covers what the skill does, how to install it, how to trigger it, what it needs, and the limits to keep in mind.
CTO Advisoralirezarezvani/claude-skills · MIT Download the skill Quick start
- 1Download the ZIP. Download the cto-advisor ZIP from this page. The licence is MIT.
- 2Unzip into a skills folder. Unzip into ~/.claude/skills/ for all projects, or into a project's .claude/skills/ for one project.
- 3Or upload in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the ZIP.
- 4Check the folder structure. Confirm SKILL.md, references/ and scripts/ are present together in the cto-advisor folder.
- 5Ask a matching question. Describe a tech debt, team scaling, architecture decision or metrics task in plain language, with your real context.
- 6Run the bundled scripts if needed. Install Python 3 and run the scripts from the skill folder, using your own inputs.
- 7Validate the output. Check estimates and scores against your own data before sharing them with stakeholders.
What is an agent skill and what does CTO Advisor do?
An agent skill is a folder containing a SKILL.md file with instructions, plus optional reference documents and scripts. The agent reads the skill when your request matches its description, so you do not have to paste the frameworks into every chat.
The cto-advisor skill gives the agent frameworks for technology strategy, engineering team leadership, architecture governance, vendor and platform decisions, and crisis management. It includes step-by-step workflows for a tech debt assessment, writing an architecture decision record (ADR), and a build vs buy analysis, along with a set of key questions a CTO asks and a metrics dashboard built on DORA metrics.
- Tech debt assessment: a severity-scored inventory, a prioritized remediation plan and a debt ratio check (target below 25%)
- ADR creation: a template with context, options, decision and consequences, plus a validation checklist
- Build vs buy: a weighted scoring table with a default rule to buy unless the capability is core IP
- Engineering metrics: DORA metrics (deployment frequency, lead time, change failure rate, MTTR) with reference guidance
How do you install the CTO Advisor skill?
You can install the skill in two ways. For Claude Code, download the ZIP from this page and unzip it into your personal skills folder at ~/.claude/skills/, or into a project's .claude/skills/ folder if you want it available only in that project. For the Claude apps, download the ZIP and upload it under Settings → Capabilities → Skills.
After unzipping, the folder should contain SKILL.md, the references/ folder and the scripts/ folder. Keep that structure intact, because SKILL.md points the agent to the files in references/ and the scripts in scripts/.
- Personal (all projects): unzip into ~/.claude/skills/
- Single project: unzip into your project's .claude/skills/
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
~/.claude/skills/cto-advisor/SKILL.mdHow do you trigger the skill with natural requests?
You do not need a special command. The skill's description lists the situations it covers, so ask in plain language. Phrases that match include: assessing technical debt, scaling an engineering team, evaluating technologies, making architecture decisions, establishing engineering metrics, and any mention of CTO, tech debt, team scaling, DORA metrics or technology strategy.
Give the agent context: your team size, the systems involved, the constraints and the decision you need to make. The frameworks produce more useful output when they have real inputs.
- "Help me prepare a tech debt review for our payments and auth services."
- "We have 12 engineers and are about to hire 10 more. How should we structure the team?"
- "Draft an ADR comparing building our own job queue with buying a managed service."
- "Set up a DORA metrics baseline for a team that deploys about once a month."
What do the bundled scripts do, and what do they need?
The skill bundles two Python 3 scripts. scripts/tech_debt_analyzer.py scores a list of debt indicators you provide and can write a report with --output report.json. scripts/team_scaling_calculator.py models engineering team growth and cost. Run them from the skill folder with python scripts/tech_debt_analyzer.py or python scripts/team_scaling_calculator.py.
The analyzer does not scan your codebase. It scores the indicator lists you supply, so its output reflects your inputs. The scoring is heuristic and has not been empirically validated, and the TCO figures and team ratios are generic templates.
- Requires Python 3
- Runs scripts: yes
- Uses network: no
python scripts/tech_debt_analyzer.py --output report.jsonSafety notes for scripts and network use
The skill runs local Python scripts, so review them before you install if you work in a sensitive environment. According to the skill's listing, it does not use the network. Treat script output as a starting point for discussion rather than a verdict.
Do not use the skill's output as authoritative financial, legal or hiring advice. Replace the example TCO numbers, ratios and cost-to-fix estimates with your own data and judgement before presenting them to stakeholders.
- Read scripts/ before running them
- Feed the analyzer real inventories from your team, not guesses
- Validate estimates with the relevant tech lead
Tips for getting useful output
Ask for one artifact at a time, such as a single ADR or one remediation plan, so the agent can follow the template closely. Provide the criteria that matter to you, because build vs buy scoring depends on your weights and requirements.
Use the validation checklists the skill includes. For example, an ADR should have a 3-year TCO estimate for each option, at least one do-nothing or buy alternative, and sign-off from affected team leads.
- Name the systems and teams involved
- State your constraints: budget, timeline, team expertise
- Ask the agent to run the validation checklist before you share the result
What people use it for
- Preparing a tech debt review with a prioritized remediation plan for an engineering manager or tech lead
- Structuring and growing an engineering team as a founder or early CTO, using the senior-to-junior and manager ratios as starting points
- Drafting an architecture decision record that compares building in-house with buying a vendor
- Setting a DORA metrics baseline for a team and choosing targets for deployment frequency and MTTR
Frequently asked questions
Is the CTO Advisor skill free to use?
The skill is published under the MIT licence, as stated on this page. Check the licence file in the repository for the full terms before reuse.
What do I need to run the CTO Advisor skill?
You need an AI agent that supports skills, such as Claude Code or the Claude apps. The two bundled scripts need Python 3, but you can use the frameworks and templates without running them.
Where do I put the skill folder for Claude Code?
Unzip it into ~/.claude/skills/ for personal use across projects, or into a project's .claude/skills/ folder to limit it to that project.
Does the tech debt analyzer scan my code?
No. It scores a list of debt indicators you provide, so its output reflects your inputs. Its scoring is heuristic and not empirically validated.
Does the skill send my data over the network?
According to the listing, the skill does not use the network. It runs local Python scripts, so review the scripts/ folder before installing if your environment is sensitive.
How does this compare with just asking a chatbot about CTO topics?
The skill gives the agent fixed templates, checklists and scoring steps for specific tasks such as ADRs and build vs buy analysis. A general chat answer may not follow those structures or include the validation steps.
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
The CTO Advisor skill suits engineering managers, tech leads, founders and early CTOs who want structured help with tech debt reviews, team growth, ADRs and build vs buy decisions. It is less suitable if you need authoritative financial, legal or hiring decisions, or automated scanning of a real codebase, since the scripts score your inputs using generic templates and heuristics.
Download the skill