Post Compact — AI skill for Claude Code
Rehydrate after a MANUAL /compact.
How to install Post Compact
Installs to ~/.claude/commands/drushegh-ccmaf-post-compact.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/drushegh/CCMAF/HEAD/.claude/commands/post-compact.md -o ~/.claude/commands/drushegh-ccmaf-post-compact.md Restart Claude Code, or start a new session, for it to be picked up.
What Post Compact does
Rehydrate after a MANUAL `/compact`. A model-generated SUMMARY of the pre-compact context is now in your window — treat it as POTENTIALLY LOSSY. This is the manual counterpart to the `session-reanchor` hook (which only fires on AUTO-compaction); you run it yourself after a deliberate `/compact`.
**Step 0 — v2 self-route (read this first):** if `.claude/.framework-version` contains a `FRAMEWORK_LINE=v2` line, this project runs the CCMAF v2 plugins — invoke the `ccmaf:post-compact` skill (S
Alternatives in AI
- Model Update — 增量更新财务模型 — 用新财报/新指引/修正假设重算 DCF 内在价值、Comps 隐含价、投资逻辑与目标价,输出 before→after delta 6.6k ★
- Cwc Long Running Agents — Claude Code's built-in /goal command gives you a generator/evaluator loop out of the box: set a completion con 668 ★
- Yt Transcriber — CLI app- Give it a YouTube URL and you get a transcription with possible speaker identification and optional s 337 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Dsh Plugin Hooks
DeepSeek Harness plugin: Claude-Code-style lifecycle hooks for DSH. Configure shell commands that run automati
Aar Loop
Run an After Action Review after any AI session and write concrete lessons so your agent stops repeating mista
Repo Intel
Unified static analysis - git history, AST symbols, project metadata, doc-code sync, plus LLM-augmented file d
Secondmate Reason
Ask a read-only reasoning model (via secondmate) a hard analysis question — root-cause, triage, plan review, p
Legible
Plain writing, a live canvas and local explainer videos for Claude Code, after Karpathy's post on reading LLM
AI Research Scientist Skill
Claude Code skill: work like a 2026 AI research scientist — research discipline, validation, SOTA recipes (LLM
Related Agents
Prose Context Extractor
Pulls one work package — a ticket, or an epic with all its child tickets — from the project-issues MCP and ret
Context Harvester
Use before non-trivial work to search project knowledge (AGENTS.md, rules, docs, ADRs, notes, changelog) and r
Otgw Device Verify
Runs the OTGW on-device verify loop (build LittleFS/firmware -> flash the bench ESP32-S3 over USB -> verify th