93pct

Autocomplete for agency next steps. Like Google autocomplete but for what to do next. Given any context, returns the top concrete viable actions ranked by RO...

Security Scan

The skill's code and instructions are internally consistent with its stated purpose (ranking concrete next steps) but it reads local financial and secret files and writes a local DB, so review those data sources before enabling it.

Purpose & Capability

The skill claims to produce ranked next steps and the included script builds a prioritized stack from local state (databases, drafts, file existence). The files and checks (dev.to draft presence, GCP SA presence, dollar DB, BTC status) line up with an 'agency next steps' planner.

Instruction Scope

Runtime instructions are simple (run suggest.py, view stack, mark done). The script reads local DBs and JSON files, checks for presence of credential files, and writes/updates a local agency.db next_steps table. It does not make outbound network requests itself, but it prints URLs and suggests commands that would invoke other scripts (e.g., btc-monitor.py, deploy-dollar-v3.py).

Install Mechanism

No install spec — instruction-only with a bundled Python script. Nothing is downloaded or extracted at install time.

Credentials

The skill requests no environment variables or external credentials, but it inspects sensitive local artifacts (WORKSPACE/dollar/dollar.db, ../../../human/btc-status.json, secrets/devto-api.json, secrets/gcp-service-account.json, /root/deploy-dollar-v3.py, /tmp/agency-b64.txt). Reading these files is consistent with planning goals, but they may contain financial or secret data that could appear in outputs.

Persistence & Privilege

always:false and default invocation rules. The script creates/updates its own agency.db table (local persistence) but does not modify other skills or global configs.

Assessment

This skill appears to do what it says, but it accesses local financial and secret-related files and writes a local next_steps DB. Before installing or enabling autonomous use: (1) Review the contents of the referenced paths (WORKSPACE/dollar/dollar.db, WORKSPACE/secrets/devto-api.json, WORKSPACE/secrets/gcp-service-account.json, ../../../human/btc-status.json, /root/deploy-dollar-v3.py, /tmp/agency-b64.txt) to ensure no sensitive secrets or private keys are present or that you're comfortable exposing summary values. (2) If you want stricter isolation, run the skill in a sandboxed workspace or remove/mask sensitive files. (3) Be cautious when approving suggested job IDs/commands — some items reference other scripts (btc-monitor.py, deploy-dollar-v3.py) which you should inspect separately before executing. (4) If you need stronger assurance, request the author clarify handling of any secrets and whether any other scripts invoked by the stack make network calls or exfiltrate data.

93% — Agency Autocomplete

Doctrine

"Plan Shannon → Penn Station"

Shannon is the currency of information (Claude Shannon, entropy theory).
Penn Station is the commuter hub — where everything converges.
The plan: route all decisions through Shannon (information value) to reach the station (execution).
Low Shannon = entropy = noise = waste.
High Shannon = signal = viable next step.

Usage

python3 /root/.openclaw/workspace/skills/93pct/suggest.py          # top 5 next steps now
python3 /root/.openclaw/workspace/skills/93pct/suggest.py --stack  # show approval stack
python3 /root/.openclaw/workspace/skills/93pct/suggest.py --done <id>  # mark completed

Output format

Each suggestion includes:

• Shannon score (0-10, information density)
• Effort (minutes)
• Blocker (what stops it)
• Job ID to approve (if exec needed)
• Exact command or action

The 93% Bar

A suggestion clears 93% if:

• It is concrete (not "consider doing X")
• It is viable right now (blocker is known and solvable)
• It produces measurable output
• It costs less than it returns