Codex context guard
Skill stark-ai-de/agent-skills/incubator/skills/codex-operations/codex-context-guard
Public Agent Skills for Codex operations, Cursor operations, Claude operations, repo maintenance, skill maintenance, productivity, and engineering workflows.
npx -y skills add stark-ai-de/agent-skills --skill codex-context-guardAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Prevent Codex context-window exhaustion during long-running refactors, repo audits, migrations, debugging sessions, or tasks with large logs, many file reads, or repeated tool output. Use when Codex context is getting high, /compact may be needed, or the user asks for context-efficient workflows.
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.5 KB, as published. Nobody here has run it
Codex Context Guard
Goal
Keep a long Codex session useful by reducing unnecessary context growth, preserving decisions in a compact handoff, and choosing bounded next steps before the thread becomes hard to continue.
Trigger conditions
- The task has many files, logs, diffs, generated outputs, or repeated tool calls.
- The user asks whether to compact, summarize, or continue in a new thread.
- The agent is about to inspect broad files without a specific need.
When to use
- Use when context budget, log volume, or file-reading scope is becoming a risk.
- Use before compacting a long session or moving work to another thread.
When not to use
- Do not use for short, self-contained tasks that have no context pressure.
- Do not use as a substitute for actually validating completed work.
Inputs to inspect
- Current objective, recent user instructions, changed files, validation output, and unresolved blockers.
- Use
git status --shortandgit diff --statbefore reading large diffs.
Rules
- Prefer targeted
rg,sed,git diff --stat, and small file ranges over whole-file dumps. - Keep one bounded objective active at a time.
- Store durable state in a handoff artifact when the work will outlive the current context.
- Summarize large outputs instead of pasting them back to the user.
- Tell the user when a compaction or handoff would preserve momentum.
Workflow
- Identify the current objective, latest decision, and next blocking question.
- Replace broad exploration with a targeted search plan.
- Track changed files, validation commands, and open risks in a compact note.
- Use references only when they solve the immediate context problem.
- When context is high, prepare a handoff and recommend
/compactor a new thread.
Commands
Useful low-context commands:
git diff --stat
git status --short
rg -n "pattern" path
sed -n '1,120p' path/to/file
Safety rules
- Do not hide failed validation behind a summary.
- Do not discard unresolved blockers when compacting.
- Do not paste secrets or full large logs into handoff text.
References
Read only when needed:
references/context-budgeting.mdfor large task tactics.references/handoff-template.mdwhen writing a handoff file.
Scripts
No bundled scripts.
Output format
Return:
- Current objective
- Context risks
- What to keep in active context
- What to move to a handoff
- Recommended next bounded action
Completion criteria
- The active objective and next action are clear.
- Large context sources have been summarized or moved to a handoff.
- Validation failures and blockers remain visible.
Failure modes
- If context is already overloaded, stop broad reads and produce a compact state summary.
- If important evidence is missing, name the missing command or file instead of guessing.
Recovery behavior
If the thread is already overloaded, stop broad exploration, write a compact state summary, name the most recent verified command result, and recommend continuing from the handoff.