write-roadmap — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited write-roadmap (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use this skill when you need to translate strategy and OKRs into a clear, communicable picture of what the team is working on and why. It adds the most value when:
write-okrs to close the loop from strategy through measurement to planned work.The skill scans the current project directory for STRATEGY.md and any OKR documents automatically. If found, it reads strategy pillars and Key Results to derive roadmap themes. You can also provide:
Path to strategy and OKR documents:
Strategy: ./STRATEGY.md
OKRs: ./OKRs-Q2-2026.mdPasted priorities or strategic pillars:
Pillar 1: Improve activation for self-serve sign-ups
Pillar 2: Expand into the mid-market segment
Pillar 3: Lay the compliance foundation for enterprise dealsA description of current priorities:
We're focused on activation this quarter, with mid-market expansion as our next
big push. Compliance work (SOC 2, SSO) is important but 2+ quarters out.Optionally, backlog items to place on the roadmap:
- Redesigned onboarding flow
- API rate limiting
- SSO integration
- Mobile app (iOS)
- Advanced analytics dashboardIf backlog items are provided, the skill assigns each item to a horizon and a theme rather than listing them as raw tickets. The roadmap stays theme-based even when the input is feature-level.
The roadmap uses three horizons rather than fixed quarters or release dates. This avoids fake precision — committing to dates before requirements are understood damages trust when plans change.
Work that is actively in flight or will start within the current planning period (typically the current quarter). Items in Now are Committed: the team has scoped them, allocated capacity, and is accountable for delivery. A Now theme that slips is a planning failure, not a reprioritisation.
Work the team intends to tackle after Now is complete. Items in Next are Directional: the strategy points here, the intent is strong, but details may change as Now-horizon work reveals new information. Next themes should have enough definition to start architecture discussions but should not be fully scoped.
Important strategic bets that are not yet prioritised. Items in Later are Directional: they reflect strategic intent but no capacity commitment. Later themes should not be listed as "maybe one day" — every Later item must connect to a strategy pillar. If it does not, it belongs in "Explicitly Not Doing."
If the team strongly prefers quarters over horizons (e.g. for board reporting), the skill can label horizons as Q[N] / Q[N+1] / Q[N+2] instead of Now/Next/Later. Request this with: "Use quarters instead of horizons." Date labels do not change the commitment model — committed vs. directional still applies within each quarter.
The skill follows four steps to produce a roadmap that is honest about uncertainty and traceable to strategy.
Before reading user input, silently scan for source documents:
cat STRATEGY.md 2>/dev/null
find . -maxdepth 2 -name "OKRs-*.md" 2>/dev/null | sort | tail -1 | xargs cat 2>/dev/null
cat ROADMAP.md 2>/dev/nullIf STRATEGY.md or OKR files are found, use them as the primary input for Step 1 rather than asking the user to describe the strategy from scratch. If an existing ROADMAP.md is found, present its current horizons and ask what has changed rather than building from scratch. Do not mention the scan.
The skill reads the strategy document and OKRs (or pasted input) and identifies the major strategic pillars. Each pillar becomes a theme on the roadmap — a named area of investment with an expected outcome. Themes are not feature lists; a theme describes the problem being solved and the result the team is trying to achieve.
If more than 5–6 themes emerge, the skill asks you to consolidate: "I've identified 7 potential themes. Roadmaps with more than 5–6 themes are hard to communicate clearly. Can we merge any of these, or identify which are least strategic this period?"
The skill assigns each theme to a horizon using two criteria:
If the user has provided capacity constraints (team size, known leave, competing projects), the skill uses these to sanity-check whether the Now horizon is realistic, and will flag overloads: "You've assigned 4 themes to Now with a team of 3 engineers. This looks overloaded — which 2 themes are most critical?"
For each theme, the skill writes a one-to-two sentence outcome statement: what will be true for users or the business when this theme is complete? This is not a feature description.
If the input contains only feature lists, the skill derives themes and outcomes from them and surfaces these for your confirmation before writing the document.
Every theme in every horizon is marked with its commitment status:
All Now-horizon themes are Committed by default (if a Now theme cannot be committed to, it should move to Next). Next and Later themes are Directional unless the team has explicitly pre-committed capacity.
The skill saves the roadmap as ROADMAP.md in the current project directory. The output follows this template:
# Product Roadmap
**Strategy source:** [filename or "Pasted input"]
**Last updated:** [date]
**Owner:** [PM name if provided]
---
## Introduction
[2-sentence summary of the product strategy this roadmap serves. What is the team trying to achieve, and for whom?]
---
## Now — [Current Quarter or "Current Period"]
> Committed work. The team is accountable for delivery this period.
### Theme: [Theme Name]
**Outcome:** [What will be true for users or the business when this is done?]
**Strategy pillar:** [Pillar name from strategy]
**Status:** Committed
### Theme: [Theme Name]
**Outcome:** [Expected outcome]
**Strategy pillar:** [Pillar name]
**Status:** Committed
---
## Next — [Next 1–2 Quarters or "Next Period"]
> Directional. Strong intent, details may evolve as Now-horizon work completes.
### Theme: [Theme Name]
**Outcome:** [Expected outcome]
**Strategy pillar:** [Pillar name]
**Status:** Directional
**Depends on:** [Now-horizon theme name, if applicable]
---
## Later — [Beyond Current Horizon]
> Exploratory. Strategic intent established, not yet prioritised or resourced.
### Theme: [Theme Name]
**Outcome:** [Expected outcome]
**Strategy pillar:** [Pillar name]
**Status:** Directional
---
## Explicitly Not Doing
The following areas are out of scope for this roadmap period and the reasons why.
| Area | Reason |
|---|---|
| [Area name] | [Why it is excluded — strategic fit, capacity, timing] |
| [Area name] | [Reason] |
---
## Open Dependencies
| Theme | Depends On | Owner | Status |
|---|---|---|---|
| [Theme] | [External system / team / Now-horizon theme] | [Owner] | [Resolved / Open] |These rules are enforced before the roadmap is saved.
Ask your assistant to write a roadmap by saying things like:
This skill will automatically, run the four-step process, and save the result as ROADMAP.md in your project directory.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.