Mantis Reflect
google/mantis/mantis-reflectby google2b3bbdcbb8d2No license2.3K starsListed Oct 9, 2026Updated Oct 9, 2026Repository updated 2 days ago
Extracts learnings from execution trajectories at the end of a Mantis loop. Use to parse agent conversations, extract successes, failures, and false assumptions, and append them to workspace/learnings.jsonl. Don't use for analyzing source code or writing patches.
- 2b3bbdcbb8d2Currentcommit 2b3bbdcPublished Oct 9, 2026
Source and attribution
Source:google/mantisinmantis-reflectat commit2b3bbdc
License: No license
Content belongs to its original authors. SourceWeft indexes it from a public repository.
More from google/mantis
Mantis Summarize
Generates security-focused summaries of each repository directory to map a codebase before threat modeling and planning.
Mantis Threat Model
Synthesizes a threat model from a knowledge base of architecture and entity definitions, writing THREAT_MODEL.md.
Mantis Structural Index
Builds a content-addressed structural code index with symbols, call edges and function boundaries for a pinned source tree.
Mantis Review
Independently reviews findings and filters out false positives. Use when consolidated findings need validation against the actual source code. Don't use for reproducing crashes or patching code.
Mantis Reproduce
Generates and runs crash reproducers in isolated sandboxes to empirically verify security findings.
Mantis Plan
Formulates a defensive security review plan and roadmap for a codebase, writing workspace/plan.json.