discovery-habits-advisor
Set up a steady rhythm of customer contact that feeds decisions
Install / Use
npx skills add deanpeters/Product-Manager-Skills --skill discovery-habits-advisorInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of discovery-habits-advisor
discovery-habits-advisor scores 90/100 on our quality scale, 1287th of 4,698 Development & Engineering skills we index (top 28%).
Its SKILL.md is 14 KB long, well organised into 37 sections with 1 code example: a thorough specification that gives an agent plenty to work with.
With 7,063 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 3 months ago, so discovery-habits-advisor is actively maintained.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 88/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-11-16. Automated pattern scan on 2026-11-16. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
discovery-habits-advisor compared with similar skills
All 4 of these similar skills score higher than discovery-habits-advisor; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| discovery-habits-advisor (this skill)by deanpeters | 90 | 7.1k | 3mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 116.8k | 8d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.1k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.3k | 2d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.6k | today | MCP Server |
Frequently asked questions
- How do I install discovery-habits-advisor?
- Run
npx skills add deanpeters/Product-Manager-Skills --skill discovery-habits-advisor. The install tabs above show the steps for each supported agent. - Which AI agents does discovery-habits-advisor work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is discovery-habits-advisor safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 88/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is discovery-habits-advisor still maintained?
- The repository was last updated about 3 months ago, so discovery-habits-advisor is actively maintained.
Skill content
View source on GitHubname: discovery-habits-advisor argument-hint: "[your team, how it talks to customers today, and what gets in the way]" description: "Set up a steady rhythm of customer contact that feeds decisions. Use when discovery happens in bursts, or never, and you want a habit the team will actually keep." intent: >- Guided triage from how a team talks to customers today to a sustainable cadence. Diagnoses the real blocker (access, time, ownership, or what happens to the evidence), fits a weekly or fortnightly touchpoint rhythm to the team and customer reach, and defines who joins, how evidence is captured, and how it reaches a decision. Right-sized so a small team starts with one habit. type: interactive theme: discovery-research best_for:
- "Moving from occasional research projects to steady customer contact"
- "Diagnosing why a team that 'knows it should' talk to customers doesn't"
- "Designing a cadence that fits long sales cycles, limited access, or a regulated customer"
- "Making sure evidence from conversations actually changes a decision" scenarios:
- "We do a research sprint once a year and then go quiet. How do we make it continuous?"
- "We can't reach end customers, only through partners. What cadence works?"
- "Our team is three people. What's the smallest discovery habit worth keeping?"
- "We talk to customers but nothing changes. What's missing?" estimated_time: "10-20 min"
Discovery Habits Advisor
Purpose
Turn customer contact from an event into a habit: a rhythm of small, regular conversations whose evidence reaches the decisions the team is making that week.
Teams usually know they should talk to customers. They don't, because discovery lives in projects (with a start, an end, and a deck), and projects get squeezed by delivery pressure. A habit fits inside normal work: a few conversations a week, notes in one place, one visible question the evidence is aimed at.
The sentence to keep in your head: the point of a habit is not more customer contact, it's shorter distance between what customers say and what the team decides.
Input
Works best with: Your team (who is on it), how you talk to customers today, and what gets in the way.
Also useful: Who you can reach and how (direct, through partners, through support), the length of your sales cycle, what decisions are coming up, and where evidence currently goes.
Anything supplied with the invocation itself (text after the skill name, a pasted context dump, or
an appended ARGUMENTS: line) counts as answers already given. Use it and skip whatever it covers;
don't re-ask.
Arriving empty-handed? That works too. The skill starts by asking how customer contact works today and what's in the way, then recommends the smallest rhythm that would work. It sizes the habit to your team, never defaults to the heaviest, and lets you move it either way.
Example invocations:
Team of 4. We run a discovery sprint once a year. Want something continuous.We only reach plant engineers through partners. What cadence can we keep?
Key Concepts
The Dial: Right-Size the Habit
Not every team can do the same rhythm. Pick a level, say why in one line, and let the user move it:
| | Level 1: Light | Level 2: Standard | Level 3: Full | |---|---|---|---| | Rhythm | One conversation every two weeks | One or two a week, same slot | Weekly touchpoints with a rotating team | | Who | The PM | PM + one designer or engineer | The product trio, rotating guests | | Evidence | One shared notes page | Notes plus a weekly one-line "what we learned" | Notes, a living opportunity tree, review in planning | | Fits | Small team, long cycles, limited access | Most teams | Teams with access and a high-uncertainty roadmap |
Level 2 is the default. Never default to Level 3. A heavy cadence a small team can't keep teaches the team that discovery is a thing they fail at. If they go lighter, name what it leaves uncovered (for Level 1: fewer conversations means slower pattern detection, and fewer people hearing customers directly).
Find the Real Blocker
Four blockers look identical ("we don't have time") and need different fixes:
| Blocker | What it sounds like | The fix | |---|---|---| | Access | "We can't reach customers" | Create a standing channel: customer advisory panel, support-ride-alongs, partner introductions, a recruiting pool kept warm | | Time | "We're always in delivery" | Fixed slot, tiny scope, shared by two people; trade a status meeting for it | | Ownership | "Everyone assumes someone else does it" | Name one owner for scheduling and one for the notes | | Consequence | "We talk but nothing changes" | Tie each conversation to a named upcoming decision; review evidence in the meeting where the decision is made |
Most teams that "can't find time" actually have a consequence problem: conversations that change nothing feel optional.
Aim Every Conversation at a Decision
A standing slot with no question becomes small talk. Keep one visible question the team is trying to answer this month, tied to a decision on the calendar ("whether to fund the bracket kit in March"). Conversations then gather evidence for something, and the evidence has somewhere to go.
Access in Long Cycles and Through Partners
Where customers are hard to reach, don't pretend a weekly slot will work. Use whatever contact exists: post-sale (commissioning, support calls), partners (ride-alongs, partner calls), and trade-show or site visits as clusters of conversations. Set the habit around the capture and review rhythm even when the conversations come in bursts: a weekly 20-minute evidence review keeps the habit alive between visits.
Capture Evidence Where the Team Will See It
One shared place, simple notes, each with who, when, and what was said in their words. Keep
observation separate from interpretation (see research-synthesis).
A weekly one-line "what we learned" posted where the team already looks (the planning doc, the
channel) does more than a perfect repository nobody opens.
Share the Hearing
Evidence heard directly changes minds in a way a summary doesn't. Rotate engineers, designers, or partner managers into conversations. When one person does all the listening, everyone else has a second-hand version and argues from opinion.
Facilitation Source of Truth
Use workshop-facilitation as the interaction protocol:
heads-up, one question per turn, numbered options, and best-guess mode on request.
Default entry mode: Guided · set in pm-skills.config.yaml (library-wide default or a per-skill override); the user's own choice in the session always wins. On the first turn, offer all three modes, then open in the default without waiting for a pick: Guided asks the first question, Context dump invites the paste, Best guess asks for a one-sentence minimum and drafts with every assumption labeled.
Anti-Patterns (what this is NOT)
- Not a research method. It sets the rhythm; interviewing and synthesis live in other skills.
- Not a mandate for weekly interviews. Level 1 is a legitimate answer.
- Not a tooling recommendation. A shared page is enough.
Application
This skill asks up to 4 adaptive questions, then recommends a rhythm.
Question 1 of 4: How does customer contact work today?
Agent asks:
"How do you hear from customers right now?
- Occasional projects (a research sprint, a round of interviews, then quiet)
- Indirectly (support, sales, partners relay it)
- Regularly, but nothing changes (we talk, then deliver as planned)
- Hardly at all"
Listen for the blocker underneath. 1 → likely time or ownership. 2 → access, plus second-hand evidence. 3 → consequence. 4 → access or fear.
Question 2 of 4: What's actually in the way?
Agent asks:
"What's the main reason it isn't steadier?
- We can't easily reach customers (access)
- We're always in delivery (time)
- Nobody owns it (ownership)
- We do it, but it doesn't change what we build (consequence)"
Confirm the diagnosis. If they pick time, ask whether the conversations are tied to a decision; the real blocker is often consequence.
Question 3 of 4: Who can do this, and how many customers can you reach?
Agent asks:
"Who on the team could take part, and roughly how many customer contacts could you arrange in a month?"
Size the level: one person and a handful of contacts → Level 1. A PM plus a partner and a steady pool → Level 2. A full trio with access and a high-uncertainty roadmap → Level 3. Recommend one, say why, and invite an override either way.
Question 4 of 4: What decision is coming up?
Agent asks:
"What's the next real decision this evidence could change?"
Tie the habit to it. That decision becomes the visible question for the month and the meeting where evidence is reviewed.
Final Step: Recommend the habit
## Your Discovery Habit: [team / context]
**Level:** [1 Light / 2 Standard / 3 Full], because [one line]. A lighter level would leave
[what it leaves uncovered].
**The real blocker:** [access / time / ownership / consequence] and the fix.
### The rhythm
- **Conversations:** [how many, when, who joins]
- **Where evidence goes:** [one place, who writes the notes]
- **The visible question this month:** [tied to the decision]
- **Weekly review:** [20 minutes, what gets said, who hears it]
### How evidence reaches the decision
[Which meeting, what the evidence changes]
### Start this week
1. [one action, small]
2. [one action]
### Skills to pair
[`discovery-interview-prep`](../discovery-interview-prep/SKILL.md), [`research-synthesis`](../research-synthesis/SKILL.md), [`opportunity-solution-tree`](../opportunity-solution-tree/SKILL.md)
Examples
Example A: SaaS, a small team with consequence trouble
User: "Team of 4 at Fieldlight. We run a research sprint every spring, then go quiet. We talk to customers sometimes but roadmap decisions don't change."
Agent (Q1): → 3 (regularly, nothing changes). (Q2): → 4 (consequence). Agent (Q3): "Two of you could join, and CS can arrange about six contacts a month." → Level 2.
Recommendation: Level 2, because you have two people and a steady pool. Blocker: consequence. Fix: tie every conversation to a roadmap decision. This month's question: "Does the route-rebuild pain justify the optimizer in Q3 planning?" Two conversations a week, PM plus an engineer rotating in; notes in one doc; a one-line "what we learned" in the Monday planning thread; evidence reviewed in the planning meeting where the decision is made. A lighter level would mean the engineer never hears a customer, so arguments stay second-hand.
Example B: Industrial, access through partners
User: "We sell controllers through 8 partners and only reach plant engineers through them. Sales cycle is 8 months. A weekly slot won't work."
Agent (Q1): → 2 (indirectly). (Q2): → 1 (access). Agent (Q3): "The PM and one application engineer; maybe 6-8 contacts a month, in clusters."
Recommendation: Level 1 with a captured rhythm, because contacts come in bursts. Use commissioning visits and partner ride-alongs as the source (clusters of conversations), keep a warm pool of five plant engineers who've agreed to a call, and run a weekly 20-minute evidence review even in quiet weeks. Visible question: "Is downtime or program risk the bigger barrier?" tied to the tooling decision in March. A lighter level would leave you with second-hand partner reports, so ask partners to bring an engineer to one call a month.
Common Pitfalls
Pitfall 1: Discovery as a Project
Symptom: A sprint, a deck, then silence. Consequence: Evidence goes stal
Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
116.8kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.1kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.3kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.6k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
