Nm Memory Palace Knowledge Intake
WarnAudited by ClawScan on May 10, 2026.
Overview
This knowledge-intake skill is mostly coherent, but it includes a GitHub Discussions publishing workflow that defaults to publishing public content using the user's GitHub account.
Install only if you want an agent to help build a persistent knowledge corpus from links and files. Before using the GitHub Discussion promotion feature, verify the repository and content carefully, require explicit approval for publishing, and make sure any imported local files or external resources are safe to store long term.
Findings (4)
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.
A knowledge entry could be published or updated in GitHub Discussions under the user's account more easily than intended, potentially exposing summaries, source links, tags, or local corpus paths.
Creating or updating GitHub Discussions is a public/account-level mutation. Making publish the default means a high-impact action can proceed with implicit acceptance rather than an explicit affirmative approval.
### Step 2: Confirm Promotion (Default: Publish) ``` This entry has reached evergreen maturity. Publishing to GitHub Discussions. [Y/n] ``` Publishing is the default action.
Change the workflow to default to no, require an explicit confirmation such as 'yes, publish', and show the target repository, category, title, and body before any GitHub mutation.
If enabled, the skill can act through the user's GitHub identity for Discussion creation or updates.
The promotion workflow relies on the user's authenticated GitHub CLI session. That is purpose-aligned for GitHub Discussions, but it means actions occur with the user's repository/account privileges.
**`gh` not authenticated**: Skip with message about running `gh auth login`.
Use a minimally scoped GitHub account/token, confirm the intended repository before publishing, and document the GitHub CLI/auth requirement in metadata.
Bad or manipulated external content could become part of the user's long-term knowledge base and affect later reasoning or recommendations.
The skill intentionally turns external resources into persistent knowledge. The boundary-marker instruction is a helpful control, but stored external content can still influence future sessions if it is inaccurate or adversarial.
4. STORE → Create structured knowledge entry ... After retrieval ... wrap the content in external content boundary markers per `leyline:content-sanitization` before proceeding
Review stored entries, preserve source attribution, keep external-content boundaries, and avoid treating imported content as authoritative without verification.
Actual behavior may depend on additional installed Night Market or leyline components outside this review.
The skill's behavior depends on external configuration/protocol components that were not part of the supplied file contents. This is not suspicious by itself, but it limits what can be verified from these artifacts.
Required config paths: night-market.memory-palace-architect, night-market.digital-garden-cultivator, night-market.leyline:evaluation-framework, night-market.leyline:storage-templates, night-market.leyline:document-conversion, night-market.scribe:slop-detector
Install only trusted versions of the referenced companion components and review them separately, especially document conversion and content sanitization protocols.