Documentation
124 tools, 13 guides — and you need about three of them. Search everything at once, or pick the sentence that sounds like your problem.
Remember & recall
21 tools“I want my AI to remember what we already figured out”
The core loop: store a lesson once, get it back the next time it matters — in any editor, in any session.
learn_from_attemptsStore a lesson AFTER any non-trivial task (deploy, debug, fix, decision); future sessions recall it.
smart_recallSemantically search cached context using natural language.
recall_feedbackTell the Brain whether a recalled lesson ACTUALLY SOLVED your problem.
Start & end a session
7 tools“I want to pick up exactly where the last window left off”
One call in, one call out. The next session opens pre-briefed instead of asking you to explain the project again.
session_startSingle-call session briefing. Call this at the START of every session INSTEAD of multiple separate smart_recall/recall_best_solution calls.
session_endSave a session summary when you finish working. Records what was accomplished, files changed, and lesson count.
session_handoffSave a detailed handoff for the NEXT chat window / session.
Understand a codebase
10 tools“I want the AI to know this repository, not just my prompt”
Turn the repo itself into memory: which files belong together, what depends on what, and who has touched it before.
index_projectIndex local source files into the cachly semantic cache so AI assistants can use semantic_search to find relevant files instead of re-reading the whole codebase every time.
brain_file_mapShow what cachly knows about a list of files — experts + related lessons per file.
brain_service_mapMap everything the Brain knows about a running service or system: who operates it, which files run in it, every known failure, and every proven fix.
Share with your team
17 tools“I want what one colleague learns to reach everyone else”
A shared brain with roles and scopes: your teammate debugs it once, you recall the answer.
team_learnStore a lesson in a shared team brain so all team members benefit. Like learn_from_attempts, but REQUIRES an author name for attribution.
team_recallRecall lessons from a shared team brain, showing who learned what. Works on any shared instance (all team members using the same instance_id).
team_confirmEndorse (review-confirm) a team lesson so trusted, human-reviewed knowledge ranks above unreviewed auto-learned entries.
Predict & plan
10 tools“I want to know what usually goes wrong before I do it”
Ask the brain first: known failure patterns for this change, contradictions between lessons, and what is next on the plan.
brain_predictREAD-ONLY — no side effects, no writes, no external network calls.
brain_predict_failuresPre-deploy failure prediction with probability percentages. Given a change context (e.g.
brain_planREAD-ONLY — no side effects, no writes, no external network calls. Generative planning layer on top of the CKG: given a task you are ABOUT to do (e.g.
Cache data & LLM answers
17 tools“I want to stop paying for the same answer twice”
The managed cache underneath it all: Valkey, Redis or Dragonfly, with semantic lookup for LLM responses.
cache_getGet a value from a running cache instance by key. Returns the stored value (string or deserialized JSON object) or null if the key does not exist or has expired.
cache_setSet a key-value pair in a running cache instance. Overwrites any existing value at the key — not idempotent for new data.
cache_deletePermanently delete one or more keys from a running cache instance (uses Redis DEL).
Keep the brain healthy
14 tools“I want to check the memory is still worth trusting”
Memory rots quietly. These measure it, clean it, and keep feeding it from git and CI without anyone typing.
brain_doctorCheck the health of your AI Brain and get actionable recommendations. Reports: lesson count, context entries, last session age, open failures, quality score, effective IQ boost, stale index.
brain_hygieneAutonomously sweep and maintain your Brain — flags stale lessons as provisional, archives long-dormant ones, and resolves contradictions where success clearly dominates failure.
brain_metricsReport the three decisive Brain metrics: (1) time-to-first-recall (onboarding friction), (2) recall-lift vs.
Reach beyond your team
21 tools“I want lessons that nobody in my company has learned yet”
Publish, import and federate knowledge — including a privacy-preserving mode that shares the signal without the details.
fedbrain_searchFedBrain context-weighted search: Search the global commons, weighting results by tech-stack similarity.
fedbrain_contributeFedBrain (Layer 6): Contribute a lesson to the global Knowledge Commons with a cryptographic knowledge certificate.
fedbrain_confirmConfirm that a syndicated lesson from the global commons worked for you.
Set up & connect
7 tools“I want to get this running in my editor”
Instances, connection strings and the one-shot wizard that writes the config for Claude Code, Cursor, Copilot and Windsurf.
setup_ai_memoryOne-shot setup of the cachly 3-layer AI Memory system for a project.\n\nLayer 1 — Storage: your cachly instance (Valkey, persistent across sessions)\nLayer 2 — Tools: learn_from_attempts +…
create_instanceCreate a new managed Valkey/Redis cache instance on cachly.dev. Free tier provisions in ~30 seconds.
list_instancesList all your cachly cache instances with their status and connection details.
All guides
13 pagesThe long-form pages, when a single tool is not the answer.