10 actions in 8 pieces of software · ranked by words, meaning and measured trust · 662 ms
skillsickn33/agentic-awesome-skillsrecord
- context-compression
When agent sessions generate millions of tokens of conversation history, compression becomes mandatory. The naive approach is aggressive compression to minimize tokens per request.
- context-compression
skillsickn33/antigravity-awesome-skillsrecord
- context-compression
Design and evaluate compression strategies for long-running sessions
- context-compression
skillmuratcankoylan/Agent-Skills-for-Context-Engineeringrecord
- context-compression
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.
- context-compression
skillsickn33/agentic-awesome-skillsrecord
- context-optimization
Context optimization extends the effective capacity of limited context windows through strategic compression, masking, caching, and partitioning. The goal is not to magically increase context windows but to make better use of available capacity.
- context-optimization
- 5context-compressorfound in code, not yet checkedno passport yet
MCP serverangrysky56/context-compressor
- compress_text
Compress inline text without requiring a file on disk.
- compress_pages
Compress one or more wiki pages / carryover files into latent chunk summaries.
- compression_stats
Global compression statistics.
- compress_text
- 6context-architectfound in code, not yet checkedno passport yet
- context_compress_output
Compresses noisy tool output: collapses test logs, minimalizes git diffs, or condenses file trees.
- context_compress_output
- 7context-scope-packfound in code, not yet checkedno passport yet
MCP serverzach-source/context-scope-pack
- compress
Compress text using the SCOPE algorithm.
- compress
- 8mcp-context-managerfound in code, not yet checkedno passport yet
- compress_context
Compress the context by keeping high-importance and recent messages. target_ratio = how much of the original to keep (0.3 – 0.6 recommended). Returns tokens saved and a short summary of what was dropped.
- compress_context
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.