mcpField
D
decide

decide

github.com/alirezarezvaniProductivity29.0k380 installsUpdated 1d ago
Repository

About decide

A Claude skill that records founder-approved decisions as durable company memory, preserving dissent separately from approved memos. It is for teams using the c-level-agents decision pipeline.

Key features

  1. 01Approval check before logging
  2. 02Structured decision records
  3. 03Separate raw and approved records
  4. 04Preserves dissent verbatim
  5. 05Optional llm-wiki vault bridge
  6. 0690-day review scheduling

Use cases

  1. 01Record a boardroom decision after approval
  2. 02Retain dissent for later post-mortems
  3. 03Revisit decisions at a review checkpoint

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/alirezarezvani/claude-skills --skill decide

After installing, type/decidein Claude Code to confirm it loaded.

Details

From the project README

/c-level-agents:decide — Log the Decision

Command: /c-level-agents:decide <memo-path>

Logs the founder's decision via the decision-logger skill. This is the gate where in-session deliberation becomes durable company memory.

Pipeline Position

/c-level-agents:office-hours  →  /c-level-agents:brief  →  /c-level-agents:boardroom  →  /c-level-agents:decide  →  /c-level-agents:execute  →  /c-level-agents:post-mortem
                                                       ↑ you are here

Two-Layer Memory Model

The decision-logger skill maintains two layers:

  1. Raw transcripts — every boardroom session, every advisor's Phase 2 position, every dissent. Stored under ~/.claude/decisions/raw/. Reference only, never feeds back automatically.
  2. Approved decisions — only the founder-signed memos. Stored under ~/.claude/decisions/approved/. Feeds into future /c-level-agents:office-hours and /c-level-agents:founder-mode calls.

This split prevents the system from "remembering" unresolved debates as if they were decisions.

Input

A board memo file (output of /c-level-agents:boardroom).

Workflow

  1. Read the memo path
  2. Verify it has founder approval (status: APPROVED)
  3. Extract structured decision record:
    • Decision title
    • Date decided
    • Option chosen
    • Success + kill criteria
    • Dissent (preserved)
    • Review checkpoint date
  4. Append to ~/.claude/decisions/approved/<YYYY-MM-DD>-<slug>.md
  5. Update the raw transcript pointer
  6. If llm-wiki bridge configured, write to vault (~/company-vault/10-decisions/)
  7. Schedule auto-revisit (90 days)

Output Record Format

# Decision: <title>
**Decided:** YYYY-MM-DD
**By:** <founder name>
**Memo:** <link to boardroom memo>
**Brief:** <link to original brief>
**Review checkpoint:** YYYY-MM-DD (90d default)

## Decision
**Chose:** <option>
**Rejected:** <other options + one-line why>

## Success Criteria (binding)
- <metric, threshold, timeframe>

## Kill Criteria (binding)
- <metric, threshold, action>

## Preserved Dissent
- **<dissenter>:** <unresolved concern>
- (preserved verbatim; dissent never erased)

## Next Action
- `/c-level-agents:execute` → 90-day plan due <date>

## Status History
- YYYY-MM-DD: APPROVED

Why Preserved Dissent

The biggest risk in approved decisions is forgetting why someone disagreed. When the kill criteria trigger, the dissent often turns out to have been correct. Preserving it verbatim — not summarized — keeps the company honest at post-mortem time.

Routing

  • /c-level-agents:execute <decision> — build the 90-day plan
  • /c-level-agents:freeze <decision> <days> — lock if irreversible
  • (Auto-scheduled) /c-level-agents:post-mortem <decision> — at 90-day checkpoint

Stale-Decision Audit

cs-chief-of-staff runs a weekly stale audit:

  • Decisions > 90 days without revisit → flag for /c-level-agents:post-mortem
  • Decisions with kill criteria triggered → flag immediately
  • Decisions whose company-context.md basis has changed → flag for re-examination

Related

Version: 1.0.0

Comments

Markdown supported