Loading...
Loading...
Found 14 Skills
Use when compressing agent context, implementing conversation summarization, reducing token usage in long sessions, or asking about "context compression", "conversation history", "token optimization", "context limits", "summarization strategies"
Design and evaluate compression strategies for long-running sessions
Use when conversation context is too long, hitting token limits, or responses are degrading. Compresses history while preserving critical information using anchored summarization and probe-based validation.
This skill should be used when the user asks to "compress context", "summarize conversation history", "implement compaction", "reduce token usage", or mentions context compression, structured summarization, tokens-per-task optimization, or long-running agent sessions exceeding context limits. A core context engineering skill — also activates when the user mentions "context engineering" or "context-engineering" in the context of managing token budgets and session longevity.
Search-aware context compression workflow for agent-studio. Use pnpm hybrid search + token-saver compression, then persist distilled learnings via MemoryRecord.
This skill should be used when long-running agent sessions need context compression, structured summarization, compaction, token-per-task optimization, or durable handoff summaries that preserve decisions, files, risks, and next actions.
This skill should be used when writing, enhancing, or evaluating the launch prompt for a long-running autonomous agent or a parallel multi-agent orchestration attacking a hard problem: pseudo-formal task briefs that define terms and an exact success predicate linguistically, enumerate non-counting outcomes, set persistence rules with explicit stop and return conditions and effort floors, manage a diverse portfolio of parallel approaches with an approach registry and blocked-route bookkeeping, and gate the return on adversarial audit. Route agent topology and coordination protocols to multi-agent-patterns, runtime control surfaces and loop governance to harness-engineering, evaluator and quality-gate construction to evaluation, judge design to advanced-evaluation, and compaction or memory mechanics to context-compression and memory-systems.
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress <filepath> or "compress memory file"
Quick-reference card for all caveman modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /caveman-help, "caveman help", "what caveman commands", "how do I use caveman".
Quick-reference card for all hui modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /hui-help, "hui help", "what hui commands", "how do I use hui".
[Tooling & Meta] Compress conversation context to optimize tokens
Use this when the user explicitly requests to "write/polish NSFC grant abstract", "generate Chinese and English abstracts", or "translate Chinese abstract to English abstract". Output both Chinese and English versions: The English version must be a faithful translation of the Chinese version (no additional information, no omission of key constraints). The default limit for Chinese abstract is ≤400 characters (including punctuation), and ≤4000 characters for English abstract (including punctuation); the final limit shall prevail as specified in `skills/nsfc-abstract/config.yaml:limits`. Also output **title suggestions**: By default, provide 1 recommended title + 5 candidate titles with justifications (the quantity shall follow `config.yaml:title.title_candidates_default`). Output method: Write the results to `NSFC-ABSTRACTS.md` in the **working directory** (the file name shall follow `config.yaml:output.filename`), which includes in order `# Title Suggestions`, Chinese abstract, English abstract, and length self-check. ⚠️ Not applicable in the following cases: - The user only wants to translate a general text unrelated to grant proposals (direct translation is required instead) - The user only wants to write the main body of project justification/research content/research foundation (use the corresponding NSFC series skill instead)