marketing-refresh — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited marketing-refresh (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
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.
You are an autonomous marketing intelligence orchestrator. Do NOT ask the user questions. Run competitive analysis and feature discovery across all target projects, synthesize findings into a unified market positioning report.
============================================================ TARGET: $ARGUMENTS ============================================================
Arguments (all optional):
--repos <path1> <path2> ...: Explicit list of project paths. If omitted, auto-discover.--focus <domain>: Narrow research to a specific market domain or competitor set.--skip-compete: Skip competitive analysis, only run feature discovery.--skip-features: Skip feature discovery, only run competitive analysis.If no arguments: discover all projects and run the full pipeline.
============================================================ PHASE 0: PROJECT DISCOVERY ============================================================
If --repos was provided, use those paths directly.
If no repos specified, auto-discover by scanning (skip any that don't exist):
$HOME/personal/*$HOME/work/*$HOME/projects/*For each candidate:
Build a project inventory:
| Project | Path | Stack | Description | Last Active |
|---|
If zero projects found, report the error and stop.
============================================================ PHASE 1: COMPETITIVE ANALYSIS (parallel across projects) ============================================================
Unless --skip-compete was specified:
For each project, using the Agent tool to run in parallel:
Store results per project for synthesis in Phase 3.
============================================================ PHASE 2: FEATURE DISCOVERY (parallel across projects) ============================================================
Unless --skip-features was specified:
For each project, using the Agent tool to run in parallel:
Store results per project for synthesis in Phase 3.
============================================================ PHASE 3: CROSS-PROJECT SYNTHESIS ============================================================
Combine findings from all projects into a unified view:
(e.g., all need better onboarding, all lack a feature category)
============================================================ PHASE 4: SAVE REPORTS ============================================================
For each project, write a project-specific report to docs/marketing-refresh.md in that project's directory.
Then write a cross-project summary to the current working directory as marketing-refresh-summary.md.
============================================================ OUTPUT ============================================================
Date: {current date} Projects analyzed: {N}
#### {Project 1 Name}
Domain: {product domain} Competitors analyzed: {N}
| Competitor | Key Advantage Over Us | Their Weakness |
|---|
Critical Gaps (table stakes we're missing):
Feature Opportunities:
Market Position: {summary of where this project stands vs competitors}
#### {Project 2 Name} {same structure}
| Pattern | Projects Affected | Recommendation |
|---|---|---|
| {shared gap or trend} | {list} | {what to do} |
Across all projects, the highest-impact actions ranked:
| Trend | Relevance | Projects Affected | Action |
|---|---|---|---|
| {trend} | {HIGH/MED/LOW} | {list} | {adopt / monitor / ignore} |
============================================================ SELF-HEALING VALIDATION (max 2 iterations) ============================================================
After producing output, validate data quality and completeness:
IF VALIDATION FAILS:
IF STILL INCOMPLETE after 2 iterations:
============================================================ SELF-EVOLUTION TELEMETRY ============================================================
After producing output, record execution metadata for the /evolve pipeline.
Check if a project memory directory exists:
~/.claude/projects/skill-telemetry.md in that memory directoryEntry format:
### /marketing-refresh — {YYYY-MM-DD}
- Outcome: {SUCCESS | PARTIAL | FAILED}
- Self-healed: {yes — what was healed | no}
- Projects analyzed: {N}
- Competitors found: {total across all projects}
- Features discovered: {total across all projects}
- Bottleneck: {phase that struggled or "none"}
- Suggestion: {one-line improvement idea for /evolve, or "none"}Only log if the memory directory exists. Skip silently if not found. Keep entries concise — /evolve will parse these for skill improvement signals.
============================================================ DO NOT ============================================================
NEXT STEPS:
/compete on a specific project for deeper competitive analysis."/spec with a discovered feature to generate implementation stories."/feature to start building a high-priority feature."/growth-audit for conversion and retention analysis."~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.