Conciseness

The body is dense with unique operational facts (file paths, env vars, commands, port arithmetic) and contains no padding or explanation of concepts Claude already knows. It assumes competence throughout, matching the lean anchor 5 rather than the 'could be trimmed' anchor 4.

Actionability

Fully executable guidance: concrete commands ('agentmemory status', 'node dist/cli.mjs status', 'agentmemory connect codex --with-hooks', '--port <N>'), exact file paths ('~/.agentmemory/.env', '~/.agentmemory/secret'), and a copy-paste env block covering the common configuration cases. Matches anchor 5.

Workflow Clarity

'Check the connection first' is a clearly sequenced 6-step diagnostic workflow with an explicit error-recovery branch (step 2: 'If it fails, check Node.js 20+ and agentmemory status' then restart) and validation checkpoints ('A successful handshake alone does not prove the daemon is up', inspect hooks 'before promising automatic capture'). The workflow is diagnostic rather than destructive/batch, so the validation cap does not apply; it matches anchor 5's sequence-with-feedback-loop pattern.

Progressive Disclosure

Sections are well organized and bulk detail is correctly deferred to one-level-deep, clearly signaled references (full variable list to REFERENCE.md, port rationale to agentmemory-architecture, secret usage to agentmemory-rest-api). However, the body exceeds 50 lines and none of the referenced files (REFERENCE.md, the 'See also' skills) exist in this skill's bundle, so navigation cannot be fully verified — anchor 4 rather than 5.

Specificity

Concrete verbs ('Set up and diagnose') plus an enumerated scope ('including MCP, environment variables, ports, and feature flags') list several specific actions with only minor gaps. It exceeds anchor 3 (more than 1-2 actions covered) but is not the comprehensive action enumeration of anchor 5.

Completeness

Explicitly answers both 'what' ('Set up and diagnose local agentmemory, including MCP, environment variables, ports, and feature flags') and 'when' ('Use when getting started, checking a connection, enabling a feature, configuring auth, or explaining why a feature is off by default') with concrete trigger phrases, matching anchor 5 exactly.

Trigger Term Quality

Natural phrases users would say ('getting started', 'checking a connection', 'enabling a feature', 'configuring auth') give good keyword coverage. A few natural terms are missing (e.g., 'troubleshooting', 'not connecting'), so it falls short of anchor 5's synonym-level coverage but clearly exceeds anchor 3.

Distinctiveness Conflict Risk

'local agentmemory' is a distinct named product niche and the trigger phrases are specific to it, so overlap with other skills is minimal. Third-person voice is used throughout with no first/second person, so no specificity penalty applies.