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.
npx skills add KevinZai/commander --skill ccc-memorygit 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/skills/kevinzai/commander/ccc-memory)<a href="https://agentmods.dev/skills/kevinzai/commander/ccc-memory"><img src="https://agentmods.dev/badge/skills/kevinzai/commander/ccc-memory/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/skills/kevinzai/commander/ccc-memory"><img src="https://agentmods.dev/badge/skills/kevinzai/commander/ccc-memory.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 | $0.00055 | $0.01710 |
| Opus 5.5 | $0.00022 | $0.00684 |
| Sonnet 5.5 | $0.00011 | $0.00342 |
| Haiku 4.5 | $0.00006 | $0.00171 |
Grade A, and why
ccc-memory 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
$ccc-memory — Two-tier persistent memory
CC Commander · $ccc-memory · important facts survive the end of a session without bloating every turn's context.
On Codex, AskUserQuestion becomes a plain-text question and AGENTS.md plays the role of CLAUDE.md.
When to use / when not to
- Use after resolving a tricky bug (save the root cause), after an architectural decision (record the reasoning), when a project outgrows one CLAUDE.md section, or before a
/clearthat would lose a specific insight. - Not for session-scoped state that ends with this session: use
$ccc-save-session. - Not for searching past knowledge: use
$ccc-recall. - Not for per-skill notes: put those in the skill's own body.
The two tiers
Tier 1 — CLAUDE.md (always loaded). Project type and stack, key paths and conventions, daily commands, cross-cutting style rules. Everything here costs context on every turn, so keep it tight. Use $ccc-claudemd to audit or restructure it.
Tier 2 — memory/ (on demand). Read only the file a task needs:
| Path | Holds |
|---|---|
memory/YYYY-MM-DD-<slug>.md |
Dated learnings, decisions, corrections |
memory/decisions/<slug>.md |
One file per significant call and why it was made |
memory/gotchas/<slug>.md |
Known footguns, such as a renamed API or changed syntax |
memory/preferences.md |
Curated long-term preferences that did not fit CLAUDE.md |
If the project already uses tasks/memory/, use that root instead of memory/.
What happens
-
Locate the memory root.
ls -d memory tasks/memory 2>/dev/null; ls memory tasks/memory 2>/dev/null | head -40No output → there is no memory root yet;
addandcorrectcreatememory/withmkdir -p memory(and the needed subdirectory) before writing. -
Route by
$ARGUMENTS. No argument → ask the first decision below.read(start of a session): skim the filenames from step 1, then Read only the few that match today's task. Report which files were read.add: ask the second decision below for the kind, then write the note with the template in step 3.correct: append a dated entry tomemory/preferences.mdin corrective framing ("Do X instead of Y — check: …"), with the source of the correction.consolidate(weekly, or before big work): list notes referenced repeatedly and propose moving them into CLAUDE.md; list stale dated notes and propose moving them tomemory/archive/. Show the proposed CLAUDE.md diff and the move list, and apply nothing until the user approves with the third decision below.
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.
- 3d ago Changed · +79 lines · +14 tokens per session df3f217e5c3c
- 1mo ago First seen · 86 lines · 41 tokens per session scan A baed629b8d07
ccc-memory is a skill published in the GitHub repository KevinZai/commander (6 stars, last pushed yesterday), licensed MIT. It adds 55 tokens to every session and 1,710 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
context-hygiene
Use when starting a project, reading project documentation or history, or resolving conflicts between documents. For current state, read only the latest documents designated by the project; do not proactively read archives; do not duplicate the spec; use long-term memory only for environment, people, and preferences…
agent-memory-mcp
A hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).
report
Writes the session final report to a file, then prints only the path and a one-line summary. Fires when the prompt contains "Report per memstack:report", and also when the prompt begins with a standing trigger configured through MEMSTACKREPORTONTASKPROMPTS or MEMSTACKREPORTTRIGGERS. Dormant otherwise.
token-optimization
Use when the user says 'token optimization', 'save tokens', 'context window', 'reduce tokens', 'token stack', or 'TokenStack', or asks about extending context window capacity. Covers TokenStack, the built-in compression proxy that shrinks Claude Code tool output before it reaches the Anthropic API. Do NOT use for…
state
Use when the user says 'update state', 'project state', 'where was I', or at session start to load current context.
magic-compact
Compact this Claude Code session.