smith-ctx-claude

github:tianjianjiang/smith · first committed 2026-01-10 · source · badge

Claude Code context management with /compact and /clear commands, CLAUDE.md persistence, and Tool Search optimization. Use when operating in Claude Code IDE or when context exceeds 60%. Activate for context optimization in Claude sessions.

Skill Health: not yet scored

Listed, not yet scored. This skill has no public model test yet. It is in the scoring queue; submitted skills go first, then the most-installed.

Version history

41 commits to this file on GitHub (newest first).

2026-10-18 · commit 72d8060 · docs(ctx-claude): save through smith-checkpoint and restore from one backend (#314)
2026-10-17 · commit d1ff329 · fix(ctx-claude): require a chat preview and the user's reply before an external write (#306)
2026-10-16 · commit ca7d3fb · docs(guidance): base the consent rule on a transaction model (#304)
2026-10-16 · commit 06693c9 · fix(ctx-claude): keep external-write-guard asking, widen its tool list (#303)
2026-10-05 · commit aa6cc5c · docs: remove dated and conflicting prompt text (#292)
2026-10-02 · commit 4ffb46c · feat(skills): say when to use each skill in its description (#290)
2026-10-02 · commit 438aeb1 · feat(smith-ctx-claude): lint skill files for point-in-time content (#288)
2026-09-08 · commit d95eb33 · refactor(smith): rename plan skills to smith-mode-*, add smith-sdlc, repair the plan hooks and their tests (#258)
2026-09-02 · commit 6c51dae · fix(checkpoint): carry session content in the checkpoint body (#221)
2026-09-02 · commit 1697e7f · fix: complete skill rename cross-reference updates (#215)
2026-08-30 · commit 8d40dfb · feat(smith-ctx-claude): add tool-output-hygiene PostToolUse hook (#210)
2026-08-30 · commit 943d76e · refactor(skills): compress frontmatter descriptions for token efficiency (#209)
2026-08-30 · commit dafa86a · refactor(ctx-claude): delegate memory output to smith-checkpoint
2026-08-30 · commit d55fa1f · refactor(ctx-claude): separate hook output from agent instructions
2026-08-30 · commit f4bdb1b · refactor: establish single source of truth for context management (#197)
2026-08-29 · commit 5e39559 · feat(ctx-claude): token-based context management architecture (#194)
2026-08-29 · commit cbf081f · fix: remove auto-memory references after feature disabled (#193)
2026-07-16 · commit 296397d · feat(checkpoint): make checkpoints self-reloading via memory-restore flag (#141)
2026-07-12 · commit f856231 · docs(skills): drop XML tag skeleton for plain Markdown across all skills (#128)
2026-06-27 · commit 0f6199c · docs(skills): trim always-loaded surface ~27%, move catalog+refs to REFERENCE (#117)
2026-06-27 · commit ee656bc · docs(skills): fold feedback-memory deltas into covering skills (#116)
2026-06-25 · commit 9973039 · docs(skills): make repeatedly-forgotten rules reachable at decision time (#107)
2026-06-25 · commit 9025d41 · feat(smith-settings): new skill — settings scope map + convention-validator hook (#105)
2026-06-25 · commit 5f74fb0 · docs: adopt «guillemet» placeholder delimiter (supersedes [bracket]) (#101)
2026-06-25 · commit 38bc8bd · docs: enforce [placeholder] convention ecosystem-wide (kill angle-bracket placeholders) (#99)
2026-06-24 · commit 73a0a81 · refactor(smith-ctx-claude): split Layer-3 REFERENCE.md + 5 CC-mechanics edits (#97)
2026-05-24 · commit 7a7c35a · feat(skills): generalize 23 lessons learned into 6 skill sections (#83)
2026-05-21 · commit 12f01d8 · feat(ctx-claude): document /insights and session-report surfaces (#81)
2026-05-21 · commit 4194de2 · feat(ctx-claude,style): slash-command rule + compound branch naming (#80)
2026-05-21 · commit 83a2803 · fix(skills): rename smith-{auto-mode,browser-mcp} to underscore per smith-style (#79)
2026-05-21 · commit 815324f · fix(ctx-claude): clarify '/goal' resume wording (#76)
2026-05-21 · commit a58b8d2 · feat(ctx-claude): document claude agents, /goal, /fast 4.7, system reminders (#75)
2026-05-21 · commit de101b8 · feat(auto-mode): add smith-auto-mode skill for classifier denial recovery (#72)
2026-03-18 · commit 33a4ac2 · fix(plan-claude): revert 200K default, DRY context resolution (#67)
2026-03-10 · commit bd2f673 · feat: add hooks reference, MCP lifecycle docs, and smith-style activation (#60)
2026-03-01 · commit d98c949 · feat: add pre-operation verification and proactive testing rules (#58)
2026-02-20 · commit bb63425 · feat(plan-claude): session isolation, graduated context, skill refinements (#57)
2026-02-09 · commit 5b4c667 · feat(plan-claude): unify stop hook with real token percentage (#54)
2026-02-08 · commit e5a84e1 · refactor: unify context management - eliminate /compact for Claude Code (#53)
2026-02-07 · commit 5ef711f · feat: add guardrails from insights friction analysis (#51)
2026-01-10 · commit 81c7c35 · refactor: rename all skills with smith- prefix for Claude Code compatibility (#50)

Archived snapshots