🧠 Brain Analysis

See what feeds your agents. See what you can save.

12 agents FREE

Token Breakdown per component

33,800 total across 6 agents

What each part means

skills = task instructions the agent follows
tools = functions and APIs the agent can call
memory = user context and session history
guidance = behavioural rules and constraints
identity = the agent's role and personality

What CHISEL Can Save You LITE FULL per day

Your provider cost: $ /M input tokens 50 turns/day avg

-22%

Lite saves / turn

-35%

Full saves / turn

17.1K

Tokens saved / day

$0.52

Dollars saved / day

🔒 Full compression saves $0.83/day vs Lite's $0.52 — that's $113/year extra

Pro gives you Full compression + Auto-Watch + Token Optimiser + never-pruned history + fleet comparison

Compression see the diff, then apply

Lite: compress guidance blocks · Full: +memory culling +skill dedup +context refactor

Before compression

4,200 tok

When using tools, you MUST:

1. Call exactly one tool per turn

2. Pass all required parameters as shown

3. Do not add extra text before/after

4. Wait for the tool response before proceeding

(14 more lines...)

After (Lite compression)

3,276 tok -22%

Call one tool per turn with exact params.

No extra text. Wait for response. Report errors.

COMPUTER_USE_GUIDANCE (compressed):

Use coordinates, type text, take screenshots

to verify. Return to chat when done.

PLATFORM_HINTS (shared constant):

Use MEDIA_DELIVERY_HINT string for files.

(9 blocks compressed, ~924 tokens saved)

Step 1: Preview

See what changes before applying. The diff above is what you get — no file is modified.

Step 2: Apply

Write the compressed version to your agent's SOUL.md file. A backup is created automatically.

🧠 Token Optimiser PRO learns from 200+ turns

Beyond simple compression. The Optimiser analyses every turn your agents take — which skills are used, which rules trigger, which memory gets referenced — then surgically removes what's unused and restructures what remains.

🔒

📊 Learning progress

58% — learned from 116 turns Goal: 200 turns

Skills never triggered3 of 8

Guidance rules stale2 of 5

Memory sections unused— (insufficient data)

At 200 turns, Optimiser will recommend what to prune — projected 8-12% additional savings on top of compression.

📈 Projected savings

-22%

Lite (current)

-35%

Full (available)

-43% to -47%

Projected with Optimiser after 200 turns

Lite compression + Optimiser pruning = deeper savings

→ At 50 turns/day, you'll have enough data in ~2 days

🔒 Unlock Token Optimiser + Auto-Watch + Full compression with Pro →

Drift & Usage 7-day trend

Component drift

Per-turn token usage last 24h FREE

24h ago Now

🔒 Pro unlocks never-pruned history & fleet-wide comparison →

🔓 FREE

CHISEL Lite

• Compress guidance blocks
• Per-agent breakdown & drift
• 24h per-turn timeline
• 7-day component trends

22% savings per turn

🔒 PRO — $9/mo Solo · $49/mo Team

CHISEL Full + Optimiser

• Full compression (memory + skills + context)
• Auto-Watch daemon
• Token Optimiser (learns from 200 turns)
• Never-pruned history & fleet comparison
• Cumulative fleet savings dashboard

Upgrade → Up to 47% savings + $113/year back