Harny Action Record
Server Details
A record of every action your agent takes for each person: what, for whom, when, with what result.
- Status
- Healthy
- Uptime
- 90.5% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- harny-ai/skills
- GitHub Stars
- 0
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 1 tool
With only one tool in the set, there is no possibility of selecting the wrong one. The description is dense (memory, connected accounts, actions, action record, key issuance) but the action itself is unambiguous.
`provision_app` follows a clear verb_noun snake_case convention, which matches the example patterns. With a single tool there is nothing to be inconsistent with, though the pattern is a sensible one to extend.
A one-tool server is thin even for a bootstrap purpose; provisioning is inherently a one-way operation but leaving the surface at a single call gives no room for related lifecycle operations. It sits at the low end of the 'too few' band rather than being a trivial no-op tool.
The surface covers app creation and key issuance only. Once provisioned there is no way to inspect an app, rotate or revoke its key, or list connected accounts/actions via this server, so agents hit a dead end after the initial setup.
Available Tools
1 toolprovision_appProvision AppAInspect
Set up Harny for an app with no account: memory, connected accounts, actions and the action record behind one key; makes a new app and its key, as POST /v1/auth/provision does. The answer carries the app key once, and the assistant is told to put it in the owner's secret store and nowhere else. Needs no key.
Inputs: app_name (required), customers, objective, model_preset, +2 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Per-user memory, connected accounts, permitted actions and the action record for your agents.
One private feed for you and your AI agents. Review reported actions with permission.
Your agents' personal cloud database: records and logs every agent you connect can save and read.
Emails sent. Purchases made. Code shipped. See what your AI agents did in one private feed. Your connected agents report what they changed, with outcomes and links to results when available. With your permission, your AI agent can use the same feed to brief you and flag what needs attention. Reports come from your connected agents. Actions they do not report will not appear. Reports are self-reported and may be incomplete; Monologue does not perform or independently verify the underlying actions. Browsing, planning, unsent drafts and internal thoughts stay out. Keep passwords, API keys and unrelated private data out of reports.
Related MCP Servers
- AlicenseAqualityAmaintenanceAuditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.39122Apache 2.0
- AlicenseAqualityDmaintenanceEvery agent action is recorded in a SHA-256 hash chain. Prove to clients that your agent did what it said it did. Record, query, verify, and export agent activity.340 npm3MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to report the actions they take — emails sent, forms submitted, purchases made, code pushed — into a single private timeline, and optionally read back that timeline with filtering and pagination. Write and read access are granted separately through OAuth scopes, so agents can log their own changes without seeing other agents' reports.13 npmMIT
- AlicenseNot gradedqualityCmaintenanceEnables agents to search people, retrieve evidence-backed facts and timelines, prepare briefs, rank reconnect opportunities, and propose outreach while keeping writes human-approved.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.