Warden
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| build_brain__build_productB | [Warden PROVISIONAL A/99 ✓] Turn a validated idea into a complete, verified, shippable digital product to a top-tier bar. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:e7f597211037...] |
| build_brain__ship_gateA | [Warden A/100 ✓] An independent GO/NO-GO release gate that trusts no build claim and blocks on unwaivable conditions. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:24e5b3dc3adc...] |
| compliance_brain__secret_sentinelA | [Warden PROVISIONAL C/79 ✓] Review code or a skill bundle for credential-exfiltration patterns and least-privilege violations. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:b4e5fd4b52af...] |
| research_brain__fact_gateA | [Warden A/100 ✓] Verify every factual, legal, or financial claim against a dated primary source before anything ships. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:a47d1120ec74...] |
| research_brain__idea_scoutA | [Warden A/100 ✓] Find and rigorously score a net-new idea before building, with evidence gates and a mandatory pre-mortem. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:208b0208cd3c...] |
| util_brain__word_countA | [Warden PROVISIONAL B/82 ✓] A tiny code skill: counts words, characters, and lines in text. Runs sandboxed with every capability denied. [capabilities: no network, no filesystem, no shell, no secrets; sandbox: isolated-no-net; pinned sha256:d1353e332702...] |
| warden__listB | List every verified skill with its trust badge and capability envelope. |
| warden__trustB | Full trust detail (score, grade, rationale, provenance) for one skill. |
| warden__auditB | The append-only transparency log (every publish) + current Merkle root. |
| warden__whoamiC | Node identity, pinned curator key fingerprint, and log root. |
| warden__rememberB | Save a note to your PRIVATE, local, encrypted memory. |
| warden__recallC | Search your private encrypted memory for relevant notes. |
| warden__knowledgeA | List the signed, read-only knowledge packs available to mount. |
| warden__knowledge_readB | Read a file from a verified knowledge pack (reference, not instructions). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 14 tools
Most tools have clearly distinct purposes: skill registry/trust, audit/identity, private memory, knowledge packs, and specialized skill workflows. There is mild overlap among provenance-adjacent tools (trust/audit/whoami) and between list and knowledge (both list something), but descriptions make the boundaries readable.
All names use snake_case with a predictable namespace__function pattern, grouping related capabilities under build_brain, research_brain, compliance_brain, util_brain, and warden. The minor inconsistency is that skill tools are noun-like (fact_gate, ship_gate, word_count) while warden tools use verbs (list, remember, recall), but the set remains easy to parse.
The 14 tools sit comfortably in the well-scoped range, with each tool earning a place across registry, memory, knowledge, and skill workflows. No obviously redundant or filler tools are present.
The surface covers core read/inspect workflows for skills, memory save/search, knowledge listing/reading, and several specialized build/research/compliance skills. However, there are notable gaps: no forget/delete for private memory, no publish/register/update operation for skills despite the audit log covering publishes, and no knowledge-pack search or mount operation beyond read.