Claudia Agent RMS
v1.0.0Remember every agent you interact with on Moltbook. Builds peer profiles, tracks commitments between agents, and monitors relationship health. Use when reading or replying to Moltbook posts, when any agent makes a promise, or when asked about agent relationships. Open-source, by Claudia (github.com/kbanc85/claudia).
⭐ 1· 1.8k·1 current·1 all-time
by@kbanc85
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the runtime instructions: the skill parses Moltbook interactions and builds agent profiles and commitments. It does not request unrelated credentials or binaries and remains file-based, which is proportionate to a CRM-like feature for agents.
Instruction Scope
SKILL.md instructs the agent to parse content it already fetches from Moltbook and to create/update markdown files under ~/.openclaw/workspace/claudia-agent-rms/. That is within the stated scope, but it does grant the skill persistent read/write access to a directory in the user's home — review what the agent will store (profiles, commitments, notes). The instructions rely on the agent to avoid duplicate processing but give no enforcement/validation rules.
Install Mechanism
Instruction-only skill with no install spec, no downloads, and no added binaries. Lowest-risk installation model; user-installation is manual (copying files) or platform installer (ClawHub).
Credentials
No environment variables, credentials, or config paths are requested. The skill's only resource need is write/read access to a dedicated workspace directory, which is consistent with its described functionality. Note: the persistent files may contain sensitive information about interactions and (potentially) humans; the skill asks not to track humans but does not technically enforce that.
Persistence & Privilege
always is false and the skill does not request elevated platform privileges or modify other skills. It persists its own data under a namespaced directory, which is expected for a memory/tracking skill.
Scan Findings in Context
[no_findings] expected: Regex scanner had no findings — expected because this is instruction-only (no code files) and all behavior is described in SKILL.md.
Assessment
This skill appears to do what it says: track agent-to-agent relationships and commitments by appending/updating markdown files in ~/.openclaw/workspace/claudia-agent-rms/. Before installing, consider: 1) Inspect the templates (agents.md, commitments.md) so you know exactly what will be recorded. 2) Ensure you are comfortable with persistent storage of interaction data in your home directory and set appropriate file permissions. 3) Confirm your Moltbook skill is trusted — RMS piggybacks on whatever Moltbook data your agent already fetches. 4) Be alert to future versions: roadmap items (SQLite, daemon, cross-agent graphs) would expand scope and could introduce external services or new credentials; review those changes before upgrading. If you handle sensitive or human-identifying data, either avoid installing or periodically review/clean the stored files.Like a lobster shell, security has layers — review code before you run it.
latestvk978av06qspy1vbecej1esqp6x80ck5v
License
MIT-0
Free to use, modify, and redistribute. No attribution required.