$ timeahead_
← mcp scores
75
Credence — Epistemic Guard
githubgithub ↗hybrid

Epistemic guard for AI coding: tracks uncertain values and blocks writes until verified.

install
source-only — no published npm / pypi package detected.
clone and follow the build instructions in the repo: github.com/Lakshmi-Chakradhar-Vijayarao/credence-ai
owner of this server? claim your listing to get a verified badgeclaim →
score breakdown
security (35%)100
freshness (25%)100
adoption (20%)0
quality (10%)100
trust (10%)50
capabilities · what this server can do
10 tools detected·auth: API key
high risk
● active   ○ not requested  ·  hover each badge for details
fs read fs write network exec eval secrets
why high risk: fs read + fs write + network + exec + eval + secrets active — can execute code, access credentials, and make external network calls.
  • credence_registerRegister an uncertain value or claim from conversation
  • credence_verifyMark a constraint as verified with evidence
  • credence_gateCheck whether a planned write/execute action is safe to proceed
  • credence_autoverifyAuto-verify constraints matched in a user message
  • credence_constraintsList all unverified constraints for a session
+5 more tools
  • credence_scanAnnotate generated code with unverified value warnings
  • credence_self_probeExtract and register domain-relevant values from generated code
  • credence_session_summaryPlain-English digest of all unverified constraints for agent handoffs
  • credence_memory_snapshotPersist unverified constraints across session boundaries
  • credence_memory_recallInject prior-session unverified constraints into current context
raw data
weekly downloads0
github stars0
forks1
open issues0
license✓ present
readme length11240 chars
last publish0d ago
last commit0d ago
last updated10h ago
score drop alerts
get notified by email when this server's score drops 5+ points