bi-competitive-intel — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited bi-competitive-intel (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.
Monitors for signals that customers may be evaluating or being approached by competitors. Part of the Book Intelligence detection layer -- surfaces early warning so the CSM can intervene before a formal evaluation process crystallises.
This is a detection and alerting skill. It identifies competitive signals and surfaces them with evidence. It never contacts the customer about competitive concerns or initiates win-back plays. The response strategy is entirely human territory -- it depends on relationship context, account value, and competitive positioning that the agent cannot assess.
Maintain a registry of competitor names, products, and evaluation-related terms. The registry should be reviewed and updated quarterly.
Registry categories:
| Category | Examples | Notes |
|---|---|---|
| Direct competitors | [Company names, product names] | Primary competitors in your market |
| Adjacent competitors | [Company names expanding into your space] | Products that overlap with part of your value proposition |
| Evaluation language | "alternative," "switch," "migrate," "evaluate," "compare," "looking at options," "RFP" | Language that indicates active evaluation regardless of specific competitor |
| Migration language | "data export," "transition plan," "wind down," "offboarding," "contract termination" | Language that indicates the evaluation has already progressed to a decision |
See references/competitor-registry.md for the full registry and maintenance process.
Scan incoming data for keyword registry matches:
Support tickets:
Call transcripts:
Email content (if integration available):
Detect evaluation behaviour that does not involve explicit mentions:
| Signal | Detection Method | Strength |
|---|---|---|
| Data export activity spike | Product analytics: sudden increase in export API calls, bulk downloads, or data extraction feature usage | Medium |
| Admin/API documentation access | Product analytics: increased visits to API docs, data schema pages, migration guides | Medium |
| Decreased engagement + no clear cause | Engagement score declining with no correlated support issue, product change, or seasonal pattern | Low-Medium |
| Concurrent usage of evaluation accounts | Product analytics: new user accounts from the same company domain on a competitor's platform (only if cross-platform tracking is available) | High (but rarely available) |
| Strength Level | Criteria | Examples |
|---|---|---|
| High | Direct competitor mention in evaluative context by decision-maker | "We are evaluating [competitor] for next quarter" in an email from the VP |
| Medium | Competitor mention without evaluative context, or behavioural signal with clear pattern | Competitor mentioned in a support ticket asking about feature comparison; data export spike |
| Low | Engagement decline without explicit competitive indicators | Reduced login frequency with no correlated explanation |
A competitive signal combined with other risk signals amplifies urgency:
{
"account_id": "string",
"account_name": "string",
"alert_type": "competitive_signal",
"signal_strength": "high",
"signal_source": "support_ticket",
"evidence": {
"source_type": "support_ticket",
"source_id": "TICKET-8823",
"date": "2026-03-08",
"content_summary": "Customer asked support team how to export all historical data in a format compatible with [Competitor] import tool",
"contact_role": "Admin (IT Director)"
},
"correlated_signals": [
{ "type": "engagement_decline", "detail": "Meeting attendance dropped 40% in last 30 days" }
],
"account_context": {
"arr": 62000,
"health_score": 68,
"days_to_renewal": 112
},
"recommended_response_framework": "Direct competitive evaluation detected. Do not lead with competitive positioning. First, investigate whether there is an unmet need driving the evaluation. Address the root cause before defending against the competitor.",
"urgency": "immediate"
}Each competitive signal has a lifecycle:
Track the lifecycle to measure response effectiveness and time-to-response.
Competitive threat alert: Individual alert per signal detection (see Step 5 above).
Monthly competitive landscape report:
All competitive signals surface to the CSM immediately. This is high-judgment territory -- the response depends entirely on relationship context and competitive positioning.
| Signal Strength | Routing | Urgency |
|---|---|---|
| High (direct mention, evaluative) | Alert CSM + notify CS manager | Immediate |
| Medium (behavioural or indirect) | Alert CSM | Same day |
| Low (engagement decline only) | Include in risk digest | Next business day |
| Any strength + renewal <90 days | Alert CSM + notify CS manager + CRO flag | Immediate |
See references/human-decision-guide.md for the response framework.
Required:
Strongly recommended:
Optional:
Downstream consumers:
bi-risk-detector (competitive signals as risk input)bi-account-brief (competitive flags in account context)references/competitor-registry.md -- Full keyword registry and maintenance processreferences/human-decision-guide.md -- Response framework for competitive signals~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.