CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-system

Grida AI agent system work: `@grida/daemon` (DaemonServer, loopback HTTP perimeter, files/workspaces, secrets store, daemon discovery) and `@grida/agent` (the agent tenant: sessions, providers/BYOK, runtime/tool execution, skills discovery, prompts, tiers, sandbox hosts). Use for `packages/grida-daemon/**`, `packages/grida-ai-agent/**`, desktop sidecar protocol changes, agent chat transport, and bugs in agent state or streams. For pure Electron window, preload, menu, deep-link, or CDP work, use `desktop`.

71

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, highly actionable architecture skill with concrete verification commands and diagnostic procedures. The main opportunity is tightening a few verbose sections and making the verification feedback loops more explicit.

Suggestions

Add an explicit validate->fix->retry loop to the Verification section (e.g., 'if typecheck fails, fix and re-run before build') to strengthen workflow_clarity.

Tighten the anti-goals list and some Pointers entries, or move the longer diagnostic walkthroughs into a reference file, to improve conciseness and progressive_disclosure.

Consider extracting the detailed Common Diagnostics narratives into a references/ file and keeping SKILL.md as an overview with signaled links.

DimensionReasoningScore

Conciseness

Dense, high-signal architecture and diagnostic content that assumes Claude's competence (no basic-concept padding), though the ~300-line body has a few sections (anti-goals, some pointer lists) that could be tightened further.

4 / 5

Actionability

Provides copy-paste-ready verification commands (pnpm --filter ... typecheck && test), a read-only sqlite3 query, concrete file paths, and specific debugging procedures covering common cases like empty file lists and hung tool calls.

5 / 5

Workflow Clarity

Sequenced verification with explicit ordering ('A change to @grida/daemon requires rebuilding it before @grida/agent typechecks') and structured diagnostic flows, but it is a collection of procedures rather than a single rigid workflow with explicit validate->fix->retry feedback loops.

4 / 5

Progressive Disclosure

Well-organized into clear sections (Shape, Package Boundaries, Ownership, Diagnostics, Security, Verification, Pointers) with one-level-deep references to READMEs, docs (tool-design.md), and sibling skills; no bundle files exist, so all content is inline and appropriately structured with minor organization gaps.

4 / 5

Total

17

/

20

Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly demarcates a two-package niche and provides explicit use/skip triggers. Minor improvement possible by adding more user-natural synonyms alongside the package-path jargon.

DimensionReasoningScore

Specificity

Enumerates concrete components across both packages (DaemonServer, loopback HTTP perimeter, secrets store, sessions, providers/BYOK, runtime/tool execution, tiers, sandbox hosts) — comprehensive coverage of specific capabilities rather than vague language.

5 / 5

Completeness

Explicitly states what the skill covers (the two packages and their subsystems) and when to use it ('Use for packages/grida-daemon/**, packages/grida-ai-agent/**, desktop sidecar protocol changes...') plus an explicit exclusion ('For pure Electron... use desktop').

5 / 5

Trigger Term Quality

Includes natural phrases users would say ('agent chat transport', 'bugs in agent state or streams', 'desktop sidecar protocol changes'), though it leans on internal package-path jargon and misses a few common synonyms.

4 / 5

Distinctiveness Conflict Risk

Narrow niche (Grida daemon + agent packages) with explicit boundary guidance steering Electron/preload/menu/CDP work to the 'desktop' skill, minimizing wrong-skill triggering.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 4 suspicious

Warning

Total

15

/

16

Passed

Repository
gridaco/grida
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.