About Causal Memory
Causal Memory is an MCP server published by JingxuanC in the Developer Tools category: causal memory layer for AI agents — MCP server that records decision→outcome relationships. Survives compaction.
The repository has 102 stars and 11 forks, with the last commit 12 days ago. Recent commit activity is the strongest signal that reported issues will get attention.
Install
git clone https://github.com/JingxuanC/causal-memoryThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (17)
causal_directoryPinned in system prompt
Parameters not published.
counterfactual_queryWhen choosing between options
Parameters not published.
intervention_query**Before taking an action**
Parameters not published.
invalidate_decisionWhen a lesson is wrong
Parameters not published.
invalidate_patternWhen a mined pattern is wrong
Parameters not published.
prediction_reportPeriodic calibration check
Parameters not published.
reconstruct_lessonWhen you want the distilled lesson
Parameters not published.
record_decisionAfter acting on a decision
Parameters not published.
record_factWhen learning a stable fact
Parameters not published.
rememberAfter any meaningful exchange
Parameters not published.
Show all 17 tools
resolve_updatesAfter contradicting outcomes
Parameters not published.
search_causalBefore a non-trivial decision
Parameters not published.
search_factsWhen you need "what is" info
Parameters not published.
search_memoryWhen unsure which type
Parameters not published.
search_patternsTo recall cross-task lessons
Parameters not published.
trace_causeWhen something fails
Parameters not published.
trace_cause_chainDeep failure analysis
Parameters not published.
Ask AI
Ask AI about Causal Memory
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.