consciousness-emergence-memory
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Incorrect, sensitive, or instruction-like text stored as memory could be reused later and influence the agent’s answers.
The API can append memory content to persistent files and later format retrieved memories for context injection.
def store(self, content: str, memory_type: str, importance: Optional[float] = None, auto_append: bool = True) ... if auto_inject and formatted_results: response["injected_context"] = self._format_context_for_injection(formatted_results)
Review MEMORY.md, SESSION-STATE.md, and generated memory files regularly; avoid storing secrets; treat injected memory as context to verify, not as authoritative instructions.
A bad or misleading daily-log entry could be summarized into long-term memory and affect future work.
The digest process can promote daily-log content into persistent MEMORY.md by default when run.
def digest(self, days: int = 7, dry_run: bool = False) ... if not dry_run: self._update_memory_md(digested_items)
Run digest operations in dry-run/preview mode first when possible, and review proposed additions before relying on them.
The API may not run as packaged, or it may load an unintended local/site package module with the same name.
The manifest lists hyphenated files such as scripts/memory-retrieve.py, scripts/memory-score.py, and scripts/memory-digest.py, but this code imports underscore-named modules; with no install spec, the API may fail or import modules from the surrounding Python environment.
from memory_retrieve import MemoryRetrieve from memory_score import MemoryScorer from memory_digest import MemoryDigest
Before using memory-api.py, verify the package layout, prefer relative imports or importable module filenames, and run it in a clean environment.