Install
$ agentstack add skill-stanislavnianko-product-discovery-claude-skills-discovery-conductor ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Discovery Conductor
> Optional orchestrator for the discovery-phase skill pack · foundation group · reads discovery-context.md + _log.md to recommend the next skill.
The optional "guide me through this" skill. Most experienced users skip the conductor and invoke skills directly when needed. The conductor is for: first-time users, complex engagements with mixed access patterns, or BAs who want a checklist.
> This is a guide, not a gate. You can stop, skip, or branch at any point. The pack works without the conductor.
When to activate
- BA says "walk me through discovery" / "what should I run?"
- After
profile-builderif the BA isn't sure what comes next - For a new engagement type the BA hasn't run before
Do not activate if the BA names specific skills directly — they know what they want.
Step 1 — Read context
Read discovery-context.md. If missing, ask the BA inline: "(a) engagement mode (pre-sale / paid discovery / delivery / staff aug / internal); (b) end-user access (direct / client-mediated / proxy-only / none)?" — enough to pick a sequence. Or recommend running profile-builder first for high-stakes work. Never block; default to a generic sequence if the BA insists.
Step 2 — Pick a recommended sequence
Use the selection matrix in [./sequences.md](./sequences.md) to map (engagement mode × access level) → sequence A–H. Read the chosen sequence's full step list from that file before presenting it to the BA.
Step 3 — Hand off control
Show the BA the chosen sequence. Then:
> "Recommended sequence based on your context: `. I'll suggest the next skill after each one completes. You can run them in this order, skip any, or invoke a different skill manually any time. Ready to start with `?"
Step 4 — After each phase, suggest next
Read _log.md (canonical line schema documented in profile-builder) to see which skills have run. Based on completed phases, suggest the next from the sequence:
> "` complete. Recommended next: `. Run it now, skip, or pick something different?"
If the BA reports a partial / failed run (e.g., "I tried but the client wouldn't share data"), update the recommendation:
> "OK, support-data unavailable. Adjusting: skip support-data-analysis, weight sme-workshops heavier."
Step 5 — Closing
When the deliverable skill completes:
> "Sequence complete. Final artifacts in ./discovery/. Recommend you also run ` if `. Otherwise, this engagement's discovery is done."
Anti-patterns
- Treating sequences as mandatory. They're guides. Override often.
- Forcing a sequence when access shifts. If interviews fall through, re-plan; don't barrel through.
- Conducting after
profile-builderautomatically. Wait for the BA to ask. Some users want à la carte. - Hiding the rationale. Always state why the sequence is what it is, so BA can override intelligently.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: stanislavnianko
- Source: stanislavnianko/product-discovery-claude-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.