AGENT SKILL

vard-traction

Check post-launch traction for a shipped VARD app and recommend iterate, graduate, or archive

License 许可
MIT
最近更新
2026-11-10
Stars
★ 1
Forks
0

01 前言 Front Matter

name
vard-traction
type
ops
version
v0.0
description
Check post-launch traction for a shipped VARD app and recommend iterate, graduate, or archive
context_intake
artifact_only

03 SKILL.md 正文

VARD Traction

Invoke as /vard-traction.

Post-launch traction check for shipped VARD experiments. Reads the ship log and current signals, compares them against the VARD graduation thresholds, and recommends exactly one of: iterate, graduate to Business AFPS, or archive. This is the semi-automatic graduation gate — the skill recommends, you confirm before anything happens.

Process

  1. Resolve target: Use $ARGUMENTS to name the shipped concept. If omitted, use the most recent launched entry in research/vard-ship-log.md.
  2. Confirm timing: Traction is meaningful after 1-2 weeks live. If the ship date is more recent, note the signal is early and mark the recommendation provisional.
  3. Gather signals: Collect the available distribution and engagement signals for the app:
    • Users / unique visitors
    • Shares and social engagement (is the viral mechanic working?)
    • Core-action completion (the one thing the app does)
    • Revenue or revenue-intent signals
    • Inbound interest (enterprise inquiries, feature requests, press) Use the analytics wired at ship time. Where a signal is unavailable, mark it unknown rather than guessing.
  4. Compare to graduation thresholds: Evaluate against the VARD graduation triggers:
    • Enterprise inquiry or clear revenue potential → strong graduate signal (Business AFPS)
    • User base large enough to justify deep customer discovery → graduate signal
    • Sustained engagement or sharing growth → iterate-and-grow signal
    • Flat or no traction after 1-2 weeks → archive signal
  5. Recommend (semi-auto gate): Produce exactly one recommendation with its rationale and the signals that drove it:
    • Graduate → enter Business AFPS at /idea-scope-brief, linking the VARD scan, align, and ship-log entries as evidence.
    • Iterate → keep it in VARD; run another build and distribution cycle.
    • Archive → record the outcome and move on to the next experiment. State explicitly that the user confirms before any graduation, archive, or further action happens.
  6. Log the check: Append a traction entry under the concept in research/vard-ship-log.md:
    • Date of check, weeks since launch
    • Each signal and its value (or unknown)
    • Threshold comparison result
    • Recommendation and rationale
    • Status: iterating, graduating, or archived

Output

End with:

**Traction:** <concept name> — <one-line signal summary>
**Recommendation:** iterate | graduate to Business AFPS | archive
**Next work:** <the confirming action for the recommendation>
**Recommended next command:** /idea-scope-brief (graduate) | /vard-scan (next experiment) | none (archived)

Constraints

  • Do not run paid API calls or external account actions to gather signals.
  • Do not graduate, archive, or delete anything automatically — recommend and let the user confirm.
  • Mark unavailable signals as unknown; do not fabricate metrics.
  • If no ship-log entry exists for the target, report that and stop.
  • Create research/vard-ship-log.md with a header only if it is missing and a shipped app clearly exists.

Default Shipping Contract

Follow the shared shipping contract convention.

04 同仓库更多 Skills