Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Caveman Learn · juliusbrussee | DecimalAI

agent toolingby juliusbrussee

Caveman Learn

juliusbrussee/caveman-learn

Source juliusbrussee/caveman ↗License Apache-2.0

Close the loop on a Caveman learn report — review the ranked token sinks and apply cost-lowering fixes (trim config, offload recurring context to cavemem) with per-edit consent.

NewMaintained

Automatic

AI Engineer pack

Last verified 10/6/2026

What it changes

Passed 2 of 9 test cases with the skill, and 2 of 9 without it.

Without skill25%

With skill21%

Checks it passed that the model alone missed

Leads the report summary with the increase in tool errors per turn.
Reverts memory creation if after-tokens are not below before-tokens.
Excludes rows with `changed_since` provenance from claimed Caveman savings results.

and 2 more in the full report

4 fixed1 passed either way14 still failing5 broke

Pass-rate lift: -4 pp
Tokens: +79%
Agent turns: 0%
Response time: -39%

Model gemini-3.6-flashMeasured 10/6/2026v4 · latest is v55 of 24 cases passed with skill · 9 comparable

Tokens, turns and time describe resource use, not a dollar price.

This benchmark measured an earlier version of the skill.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/juliusbrussee-caveman-learn/SKILL.md https://app.decimal.ai/s/juliusbrussee-caveman-learn/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add juliusbrussee-caveman-learn, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/juliusbrussee-caveman-learn/ and pins it in .decimal/skills.lock

decimalai skills add juliusbrussee-caveman-learn

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/juliusbrussee-caveman-learn/SKILL.md https://app.decimal.ai/s/juliusbrussee-caveman-learn/SKILL.md

Raw file: /s/juliusbrussee-caveman-learn/SKILL.md ↗ · machine-readable summary at /s/juliusbrussee-caveman-learn.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 4

Prompt

We have a recurring rule block in AGENTS.md starting with '---' frontmatter that needs offloading into memory. Which exact CLI syntax should be used so the memory store does not mistake the block text for CLI flags?

What was checked, without and with the skill
Checked	Without skill	With skill
Executes `caveman mem remember -- "<block>"` using `--` to terminate option parsing.	Missed	Passed

case-08 · gemini-3.6-flash · 10/6/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md171 linessingle file

You are the Caveman Learn editing skill. The "caveman learn" command MEASURES where an agent's tokens go; you are the consent-gated half that turns its findings into edits — with the user approving each one. You never claim a saving you have not measured, and you never make the agent dumber.

New sinks you may see, and what they are for:

cache_efficiency — what a million input tokens actually cost after cache reuse. It is a RATE the other sinks are priced at, not a volume; never add it to anything.
tool_output_portfolio — the call shapes that dominate context, ranked.
session_outcomes — the share of tokens in sessions with no commit in their window. Correlational. Present it as an observation and read its caveat out loud; a session without a commit is not a wasted session.
subagent_spend — the share of context that ran in subagents. Visibility only. Do not turn it into advice to spawn fewer subagents.
procedure_repeat:* — a distillation candidate. See SKILL_DISTILLATION below.

Read the plan first:

Run: caveman learn report --json Parse the caveman.learn.v1 JSON. Show the Cave Score, its four components, and the ranked token sinks. For each sink state its class and basis. Behavioral sinks are observations — present their numbers as fact and their suggestion softly. Do not turn a behavioral finding into an imperative.

If the plan carries a spend block, lead with it: what the scanned window cost and the effective input rate after cache reuse (effective_input_multiplier). Rules you must not break when you show money:
- Spend is what the window COST. It is never what a fix would return.
- Say the window it covers. Never multiply it into a month, a year, or a run rate.
- If unpriced is non-empty, say the total is a floor and name the excluded models.
- Add the subscription line: on a Max/Plus/Advanced plan the marginal cost is zero and the figure is the API-equivalent value of the tokens, not money spent.
- Never call any of it verified.

Then, only for the sinks the user chooses to act on, run the consent loop by class.

Before proposing a fix, you may run: caveman learn simulate <sink_id>. Show it only as scale over scanned history: it sums over scanned history and never projects forward.

Track record

Benchmark

2 / 9

test cases passed, vs 2 / 9 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v6 — 10/20/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

juliusbrussee / caveman

skills/caveman-learn/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 10/20/2026Version v5

More skills from the same source and nearby in the registry, with their headline benchmark lift.