Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/KevinZai/commanderWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/kevinzai/commander/ccc-learn-eval)<a href="https://agentmods.dev/commands/kevinzai/commander/ccc-learn-eval"><img src="https://agentmods.dev/badge/commands/kevinzai/commander/ccc-learn-eval/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/kevinzai/commander/ccc-learn-eval"><img src="https://agentmods.dev/badge/commands/kevinzai/commander/ccc-learn-eval.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00053 | $0.01352 |
| Opus 5.5 | $0.00021 | $0.00541 |
| Sonnet 5.5 | $0.00011 | $0.00270 |
| Haiku 4.5 | $0.00005 | $0.00135 |
Grade A, and why
ccc-learn-eval scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 14d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
94% identical to learn-eval — 168 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ccc-learn-eval — extract, evaluate, then save
CC Commander · /ccc-learn-eval · a learned-skill draft that passes an overlap check and a verdict before it is written.
What to extract
Look for:
- Error resolution patterns — root cause, fix, and why it recurs.
- Debugging techniques — non-obvious steps or tool combinations.
- Workarounds — library quirks, API limits, version-specific fixes.
- Project-specific patterns — conventions, architecture decisions, integration patterns.
Skip trivial fixes (typos, simple syntax errors) and one-time issues (a specific outage). One pattern per skill.
What happens
-
Pick the insight. Review the session and choose the single most reusable pattern. If none qualifies, say so and stop.
-
Choose the scope. Ask with AskUserQuestion (plain-text question where unavailable), recommending the scope that fits:
- 🟢 Global —
~/.claude/skills/<pattern-name>/SKILL.md; useful in two or more projects (shell compatibility, model API behavior, debugging techniques) - Project —
.claude/skills/<pattern-name>/SKILL.mdin this repo; specific to this codebase's config or architecture
Use the directory form exactly:
<name>/SKILL.mdis what Claude Code discovers as a skill; a flat<name>.mdfile is not. - 🟢 Global —
-
Validate the target.
pattern-nameis a lowercase hyphenated slug with no/or... Check whether it exists:test -e "<skill-root>/<pattern-name>/SKILL.md" && echo exists || echo freeIf it exists, prefer the Absorb verdict below or pick a new name; overwriting needs an explicit answer in step 6.
-
Draft the skill. Redact secrets, personal data and credentials from anything copied out of the session.
--- name: pattern-name description: "What it solves, under 130 characters. Use when <trigger>." user-invocable: false --- # <Descriptive pattern name> **Extracted:** <date> **Context:** <when this applies> ## Problem <the specific problem> ## Solution <the pattern, with the commands or code that apply it> ## When to use <trigger conditions>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 14d ago Changed · +9 lines · +28 tokens per session 8f28068b35f0
- 1mo ago First seen · 117 lines · 25 tokens per session scan A a1203082d199
ccc-learn-eval is a command published in the GitHub repository KevinZai/commander (6 stars, last pushed today), licensed MIT. It adds 53 tokens to every session and 1,352 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 94% identical to learn-eval, differing in 168 lines, and is treated as a copy.
Other commands, from other repositories
context-stats
Display context window usage and token statistics.
dashboard
Compact visual progress dashboard with progress bar, task table, dependency graph, and active/blocked/next status.
compact-context
A command that saves important project status and decisions, then creates a shorter summary of a long coding-agent conversation.
rnb
RNB — opens the RuvNet Brain Console (same console as /rnbc, /rvbc, /brain-console and /ruvnet-brain:configure — every spelling works). Your whole RuvNet stack on one page. Read-only until you click.
justone
Start a Just One Prompt Challenge — ship a runnable project from a single prompt.
thread
Persistent context threads for cross-session work that spans phases. Lighter than pause/resume.