dev
278ecca018
Keeps credentials, company names, personal names and PII out of the model's context. Enforcement lives in Claude Code hooks rather than in instructions to the agent: a skill alone cannot protect anything, because by the time the agent reads a rule the surrounding context has already been sent. Credentials are removed irreversibly and marked. Entities from a user-supplied dictionary become stable aliases, rewritten back to real values on their way to disk and to the shell, so code and commands referring to them still work. Published from a clean tree; development history is not included.
34 lines
890 B
Makefile
34 lines
890 B
Makefile
CTXGUARD := python3 plugins/ctxguard/scripts/ctxguard.py
|
|
|
|
.PHONY: help test verify leaks scan status lint all
|
|
|
|
help:
|
|
@echo "test - engine unit tests"
|
|
@echo "verify - canary corpus through the real hook entry points"
|
|
@echo "leaks - scan transcripts for values that actually reached the model"
|
|
@echo "scan - report sensitive values in this repo (counts only)"
|
|
@echo "status - show what ctxguard has active"
|
|
@echo "lint - byte-compile every script"
|
|
@echo "all - lint + test + verify"
|
|
|
|
test:
|
|
python3 plugins/ctxguard/tests/unit/test_engine.py
|
|
|
|
verify:
|
|
$(CTXGUARD) verify
|
|
|
|
leaks:
|
|
$(CTXGUARD) scan-transcript
|
|
|
|
scan:
|
|
$(CTXGUARD) scan .
|
|
|
|
status:
|
|
$(CTXGUARD) status
|
|
|
|
lint:
|
|
python3 -m compileall -q plugins/ctxguard/scripts
|
|
python3 -c "import json,glob,sys; [json.load(open(f)) for f in glob.glob('**/*.json', recursive=True)]; print('json ok')"
|
|
|
|
all: lint test verify
|