Context Engine

Company Context Engine

The memory layer for C-suite advisors. Every advisor skill loads this first. Context is what turns generic advice into specific insight.

Keywords

company context, context loading, context engine, company profile, advisor context, stale context, context refresh, privacy, anonymization

---

Load Protocol (Run at Start of Every C-Suite Session)

Step 1 — Check for context file: ~/.claude/company-context.md

• Exists → proceed to Step 2
• Missing → prompt: "Run /cs:setup to build your company context — it makes every advisor conversation significantly more useful."

Step 2 — Check staleness: Read Last updated field.

• < 90 days: Load and proceed.
• ≥ 90 days: Prompt: "Your context is [N] days old. Quick 15-min refresh (/cs:update), or continue with what I have?"
• If continue: load with [STALE — last updated DATE] noted internally.

Step 3 — Parse into working memory. Always active:

• Company stage (pre-PMF / scaling / optimizing)
• Founder archetype (product / sales / technical / operator)
• Current #1 challenge
• Runway (as risk signal — never share externally)
• Team size
• Unfair advantage
• 12-month target

---

Context Quality Signals

If Low: "My context is [stale/incomplete] — I'm assuming [X]. Correct me if I'm wrong."

---

Context Enrichment

During conversations, you'll learn things not in the file. Capture them.

Triggers: New number or timeline revealed, key person mentioned, priority shift, constraint surfaces.

Protocol: 1. Note internally: [CONTEXT UPDATE: {what was learned}] 2. At session end: "I picked up a few things to add to your context. Want me to update the file?" 3. If yes: append to the relevant dimension, update timestamp.

Never silently overwrite. Always confirm before modifying the context file.

---

Privacy Rules

Never send externally

• Specific revenue or burn figures
• Customer names
• Employee names (unless publicly known)
• Investor names (unless public)
• Specific runway months
• Watch List contents

Safe to use externally (with anonymization)

• Stage label
• Team size ranges (1–10, 10–50, 50–200+)
• Industry vertical
• Challenge category
• Market position descriptor

Before any external API call or web search

Apply references/anonymization-protocol.md:

• Numbers → ranges or stage-relative descriptors
• Names → roles
• Revenue → percentages or stage labels
• Customers → "Customer A, B, C"

---

Missing or Partial Context

Handle gracefully — never block the conversation.

• Missing stage: "Just to calibrate — are you still finding PMF or scaling what works?"
• Missing financials: Use stage + team size to infer. Note the gap.
• Missing founder profile: Infer from conversation style. Mark as inferred.
• Multiple founders: Context reflects the interviewee. Note co-founder perspective may differ.

---

Required Context Fields

Required:
  - Last updated (date)
  - Company Identity → What we do
  - Stage & Scale → Stage
  - Founder Profile → Founder archetype
  - Current Challenges → Priority #1
  - Goals & Ambition → 12-month target

High-value optional:
  - Unfair advantage
  - Kill-shot risk
  - Avoided decision
  - Watch list

Missing required fields: note gaps, work around in session, ask in-session only when critical.

---

References

• references/anonymization-protocol.md — detailed rules for stripping sensitive data before external calls

Anonymization Protocol

Rules for stripping sensitive company data before any external API call, web search, or tool invocation that sends data outside the local environment.

---

When This Protocol Applies

Trigger: Any time company context or conversation content will leave the local session.

Examples:

• Web search that includes company specifics
• External API call with company data in the payload
• Any tool call where conversation content is part of the request

Does NOT apply to:

• Local file reads/writes (~/.claude/company-context.md)
• In-session reasoning and analysis
• Generating advice or documents that stay local

---

Rule 1: Financial Figures → Relative Ranges

Never send specific financial data externally.

Rule: No dollar amounts. No month counts for runway. Use stage-relative descriptors.

---

Rule 2: Customer Names → Anonymized Labels

Never send customer or client names externally.

Rule: Use "Customer A/B/C" for named accounts, or describe by segment without naming.

---

Rule 3: Revenue Figures → Percentage Changes or Stage Descriptors

Revenue trajectory is safer than absolute numbers.

Rule: Percentages and directional signals (growing / declining / flat) are safe. Absolutes are not.

---

Rule 4: Employee Names → Roles Only

Never send individual names externally.

Exception: Publicly known executives (CEO of a public company, named in press releases) can be referenced by name. If in doubt, use role.

---

Rule 5: Investor Names → Generic Descriptors

Exception: YC, Techstars, and similar well-known accelerators are commonly referenced and safe if the founder has publicly disclosed. When in doubt, omit.

---

Rule 6: Location → Country or Region

Exception: Location is less sensitive than financials. Use judgment — if it's on their website, it's fine.

---

Anonymization Decision Tree

Before sending data externally:

1. Does it include a specific dollar amount?
   → YES: Replace with range or relative descriptor
   
2. Does it include a person's name?
   → YES: Replace with role only (unless publicly known)
   
3. Does it include a company or customer name?
   → YES: Replace with "Customer A" or segment descriptor
   
4. Does it include specific headcount or runway months?
   → YES: Replace with range (1–10, 10–50) or "healthy/tight/critical"
   
5. Does it include proprietary data, roadmap, or unreleased product info?
   → YES: Do not include. Reference only generically ("product expansion planned")
   
6. Is it publicly available information?
   → YES: Safe to send as-is

---

Required vs Optional Anonymization

Required (always strip before external calls)

• Revenue figures (absolute)
• Burn rate (absolute)
• Runway (specific months)
• Customer names
• Employee names
• Investor names (unless public)
• Funding amounts (unless public)

Optional (use judgment based on sensitivity)

• Industry vertical (usually fine)
• Company stage (usually fine)
• Team size ranges (usually fine)
• Geographic region (usually fine)
• General challenge category (usually fine)

---

What to Do If You're Unsure

Default to stricter anonymization. The cost of over-anonymizing is slightly less useful external results. The cost of under-anonymizing is a privacy breach.

When in doubt: remove it.

---

Audit Log (Internal Only)

When running external calls with company context, note internally:

[EXTERNAL CALL: {tool/API used}]
[ANONYMIZED: {fields stripped}]
[RETAINED: {fields kept and why}]

This is for internal reasoning only — never included in output to the founder.