CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

raven-self-config

Viewing and changing your own settings (raven_config), mostly connecting agents and chat apps; also where to look when one of your abilities is missing or fails.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

96%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.

Exemplary instruction-only skill content: dense with concrete, copy-paste-ready commands and exact behavioral rules, clear multi-step workflows with confirmation checkpoints and error-recovery loops, and near-zero padding. The only structural refinement left is that some domain-detail sections (Sub-agents, Models) could move to reference files to shorten the always-loaded body.

DimensionReasoningScore

Conciseness

The body is lean and directive throughout — "Raven's source code, logs and config files are not a manual", "do not answer from what you believe about yourself" — with no explanation of concepts Claude already knows and every line carrying Raven-specific rules; each token earns its place.

5 / 5

Actionability

Fully executable guidance: exact tool syntax (`describe subagents.<name>`, `add subagents {"preset": "codex"}`, `set` with JSON value examples like `{"provider": "openrouter", "model": "anthropic/claude-sonnet-5"}`, `restart` with "reload"), and even a copy-paste CLI (`raven ops connection add --auth password --password-stdin --non-interactive`) covering the common cases.

5 / 5

Workflow Clarity

Multi-step flows are explicitly sequenced with validation checkpoints and feedback loops: diagnosis-before-change ("a diagnosis is not permission"), collect-all-changes → ask once → restart, and the sub-agent failure loop with a bail-out ("if three have not told you why, stop and report what you saw"), plus verification via the `set` reply's old/new values.

5 / 5

Progressive Disclosure

A single file with clear, well-ordered sections (find → change → takes effect → per-domain detail) and no nested references, but at ~215 lines the detailed per-domain sections (Sub-agents, Models) could arguably be split into one-level-deep reference files; good structure with minor organization gaps.

4 / 5

Total

19

/

20

Passed

Description

70%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 solid description: it names the tool, the domain (Raven's own configuration), and two concrete jobs (changing settings, connecting agents/chat apps), plus an explicit failure-diagnosis trigger. Its weaknesses are the second-person voice ('your own settings'), an implied rather than stated primary trigger, and thin coverage of natural synonyms users would actually say.

Suggestions

Rewrite in third person to both fix the voice penalty and sharpen specificity, e.g. 'Views and changes Raven's own settings via raven_config: connecting agents and chat apps, models, channels, and keys.'

Add an explicit 'Use when' clause covering the primary trigger, e.g. 'Use when the user asks to change, view, or connect a setting, agent, or chat app, or when a tool or sub-agent is missing or fails.'

Include more natural trigger synonyms users would say — 'configure', 'API key', 'model', and specific app names (Telegram, Slack, Codex) — to lift trigger-term coverage toward comprehensive.

DimensionReasoningScore

Specificity

Lists several concrete actions ("Viewing and changing your own settings", "connecting agents and chat apps"), matching the anchor-4 'several specific actions with minor gaps' — but the second-person "your own settings" violates the third-person voice rule, reducing the score by 1.

3 / 5

Completeness

The 'what' is clear (viewing/changing own settings via raven_config, connecting agents and chat apps) and one explicit 'when' trigger exists ("where to look when one of your abilities is missing or fails"), but the primary use case (user asks to change a setting or connect something) is only implied rather than stated, matching 'both present; when could be more explicit'.

4 / 5

Trigger Term Quality

Good natural keyword coverage: "settings", "changing", "connecting agents and chat apps", "when one of your abilities is missing or fails" — but common variations users would say ("configure", "switch model", "API key", specific app names like Telegram/Slack) are missing, so it is not comprehensive.

4 / 5

Distinctiveness Conflict Risk

Naming "raven_config" and the niche of the assistant's own settings makes it mostly distinct with a clear niche, but generic terms like "settings" and "connect" carry minor overlap risk with other admin/configuration skills.

4 / 5

Total

15

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository

EverMind-AI/Raven

Path

raven/memory_engine/skills/raven-self-config/SKILL.md

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.