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 chipfighter/coding-discipline --skill context-hygienegit clone --depth 1 https://github.com/chipfighter/coding-disciplineWrote 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/chipfighter/coding-discipline/context-hygiene)<a href="https://agentmods.dev/skills/chipfighter/coding-discipline/context-hygiene"><img src="https://agentmods.dev/badge/skills/chipfighter/coding-discipline/context-hygiene/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/chipfighter/coding-discipline/context-hygiene"><img src="https://agentmods.dev/badge/skills/chipfighter/coding-discipline/context-hygiene.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 | $0.00067 | $0.00654 |
| Opus 5.5 | $0.00027 | $0.00262 |
| Sonnet 5.5 | $0.00013 | $0.00131 |
| Haiku 4.5 | $0.00007 | $0.00065 |
Grade A, and why
context-hygiene 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 1mo 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 — 27 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Old documents can trace history, but they are not current facts. Models easily follow the most extensively documented old account instead of finding the latest, correct one. If the project guide document (Codex's AGENTS.md / Claude Code's CLAUDE.md) designates a current-state document and archive location, follow it.
What to read—and what not to read
- For current state, read only the latest document designated by the project (such as the top of
CHANGELOG.mdor a STATUS document). Do not treat an old version snapshot as current. - Do not proactively read archives or documents marked
superseded/ deprecated. When history is genuinely needed, search for and read the relevant passage rather than reading the entire archive. - When documents conflict, first follow the project's designated latest state document; if it is still unclear, ask the user. Do not choose one yourself, and never treat an archive as current.
Do not copy another set of instructions
- Point design work to the existing project guide document / architecture decision record (ADR) / current-state document instead of copying another spec—the copy will eventually go stale.
- When the reason for a direction changes, rewrite it; do not append: retaining the old rationale invites it to be repeated and pull the work in the wrong direction.
- Use memory only for facts unavailable in the repository (environment / people / preferences / lessons from past pitfalls). Read project progress from the current-state document, and record design rationale in an architecture decision record (ADR). Delete stale memory; do not archive another copy.
Project guide documents grow over time; they are not filled in up front
When the project root has no AGENTS.md (Codex) or CLAUDE.md (Claude Code), the plugin places the corresponding empty skeleton there the first time it enters the repository. It grows by confirming one thing and recording one line:
- Record only things already agreed with the user and not apparent from the code (technology choices / hard boundaries / definition of done). Do not record matters still under discussion or unsettled—filling the guide up front makes later sessions treat guesses as facts.
- Whenever such a decision is finalized, proactively ask, “Should I record this in the project guide document?” Write it only after approval, one line at a time.
- If a recorded decision later changes, edit that line; do not add a conflicting line below it (the same “rewrite, do not append” rule above).
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.
- 1mo ago First seen · 27 lines · 67 tokens per session scan A 05763d653789
context-hygiene is a skill published in the GitHub repository chipfighter/coding-discipline (25 stars, last pushed 1mo ago), licensed MIT. It adds 67 tokens to every session and 654 once invoked, about $0.0003 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
ccc-knowledge
Search, browse and add lessons in the local CC Commander knowledge base: recent lessons, topic search, category view, stats and file-edit captures. Use when the user asks what we learned, wants past lessons, the knowledge base, or types $ccc-knowledge.
ccc-handoff
Write a dense handoff file and tell the user to resume in a new chat before a long context degrades output quality. Use when a chunk of work is done, the topic is about to switch, the model starts forgetting context, or a long orchestration run is mid-flight.
ccc-memory
Keep durable project memory in two tiers: CLAUDE.md for always-loaded facts and a memory/ directory of dated notes, decisions, gotchas and preferences loaded on demand. Use when recording a decision, root cause, correction or preference future sessions must keep.
ccc-resume-session
Load a saved session file from /.claude/sessions/ (latest, by date, or by path), read it fully, and brief project, state, what not to retry, blockers and next step before touching anything. Use when resuming work, continuing where you left off, or after a fresh session.
ccc-save-session
Compress the current session into a dense reloadable file in /.claude/sessions/: decisions, files touched, what worked, what failed and why, open questions, exact next step. Use when ending a session, nearing context limits, or saying save this session.
ccc-session
Alias that routes session work to the canonical skills: save the current session with $ccc-save-session, or reload a saved one with $ccc-resume-session. Use when the user says resume session, continue where I left off, save this session, or types $ccc-session.